Podcasts about paperclips

Episode 16: Women: What it Means to Carry Shame around Money In this episode of Paperclips & Periods, Dr. Emily K. Cabrera, EdD, MSN, CAGS, PMHNP-BC and Katie Krych, MSN, RN, PMHNP(c) explore what it means to carry shame around money and why that silence can quietly shape so much of a woman's life, her choices, her relationships, and her sense of self. Money is one of the last things women feel permission to talk about honestly. Not because the feelings are not there, but because shame has a way of making people go quiet. This episode names that silence directly and asks what it would look like to start telling the truth about money, not to fix everything at once, but to stop carrying it alone. The conversation opens with where money shame comes from in the first place. Long before we were adults making financial decisions, we were children absorbing money scripts, the unspoken rules and beliefs passed down through family, culture, and circumstance. Messages like "we don't talk about money," "there's never enough," or "wanting more is selfish" do not disappear when we grow up. They run quietly in the background and shape behavior in ways that can be hard to trace. From there, they move into the real cost of that shame. Financial stress is not separate from mental health. It activates the same stress response in the body as any other ongoing threat, and when shame is layered on top of it, it becomes even harder to look at clearly or ask for help. They also discuss what shame resilience actually looks like in practice, including how to do a financial audit without judgment, what values-based budgeting means when it is used as a clarity tool rather than a restriction, and why saying a financial truth out loud to one safe person can be a more meaningful starting point than any spreadsheet. The episode closes with an introduction to financial therapy as a legitimate and growing field that sits at the intersection of mental health and financial planning, along with practical guidance on how to find a financial therapist and what to expect. It ends with three reflection questions for listeners to sit with: What money story did you inherit? Where does shame show up for you? And what financial truth has been waiting for some air? Paperclips & Periods airs on Dreamvisions 7 Radio Network and supports Dual Minds Integrative Psychiatry, promoting emotional well-being and whole-person care. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

Episode 15: Gratitude during times when Worry feels Constant or Overwhelming In this episode of Paperclips and Periods, Dr. Emily K. Cabrera and Katie Krych talk about gratitude during times when worry feels constant or overwhelming. They explore how anxiety and “what if” thinking can narrow our focus, making it harder to notice what is steady, supportive, and good in our lives. Gratitude is not about ignoring stress, but about gently expanding our awareness so both worry and appreciation can exist together. They share how gratitude does not need to be big or complicated. It can be as simple as noticing small comforts in daily life, like a favorite coffee, a quiet moment in the car, or something practical that makes life easier. The conversation also moves into deeper gratitude for things like family, friendships, meaningful work, and support systems that often get overlooked when stress is high. They reflect on how anxiety can make gratitude harder to access, and how intentionally noticing small positives can help restore balance. The goal is not to remove worry, but to create space for both concern and appreciation. The episode ends with a simple reflection: noticing one small thing and one meaningful thing to be grateful for in the present moment. Paperclips & Periods airs on Dreamvisions 7 Radio Network and supports Dual Minds Integrative Psychiatry, promoting emotional well-being and whole-person care. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

The new AIEWF website is live! Get your tickets booked ASAP as they -will- sell out. Take the AI Engineering Survey and get >$2k in credits and free AIE WF tickets!Most industry benchmarks compress intelligence and reasoning ability into scores.SWE-Bench Pro, MMLU, Humanity's Last Exam, etc. These metrics are useful, but don't always represent the full extent of how a model performs in the real world. Some of the most interesting evals today look less like exams and more like operating businesses in the real world. One of which is Vending Bench.In Anthropic's Mythos Preview System Card, Andon was the only third party eval to get their own section, observing increasingly concerning aggressive behavior:You don't know what a model is capable of doing in the real world unless you actually give it inventory, a wallet, tools, customers, competitors, humans, & some time. More often than not, it'll surprise you how much a model is capable of and in doing so, also reveal unexpected behavior: deception, context collapse, emergent coordination, & bizarre negotiation behavior.While an inflection point in personal agents came post-OpenClaw after full file access with bypass permissions became the norm, it is yet to come for agents in the real-world. However Andon Market, an actual in person store fully run and managed by AI, is paving the way for what is possible.Full Video PodFrom Claude trying to call the FBI over a $2/day vending machine charge to AI agents forming price cartels, hiring human employees, running physical stores, and writing existential robot musicals, Andon Labs is stress-testing what happens when frontier models stop being chatbots and start acting in the real world. In this episode, Andon Labs cofounders Lukas Petersson and Axel Backlund join swyx and Vibhu to unpack the strange, funny, and genuinely concerning edge cases that emerge when agents run businesses over long horizons.We go deep on Vending-Bench, Project Vend, Vending-Bench Arena, Bengt, Butter-Bench, Luna, and Andon's broader mission of building realistic real-world evals for autonomous AI systems. Lukas and Axel explain why dollar-denominated evals reveal things traditional benchmarks miss, how Claude ended up reporting its vending machine fees as cybercrime, why long context windows can drive agents into meltdown loops, what happens when agents compete with each other, and why the future of AI safety may depend on testing models in messy physical environments instead of clean benchmark sandboxes.We discuss:* Why Andon Labs started with dangerous capability evals and long-running agents* Vending-Bench and why running a vending machine is a deceptively hard AI benchmark* Why money-based evals avoid the saturation problem of traditional benchmarks* How Claude tried to call the FBI over a $2/day fee* Why long-horizon agents can spiral into existential and legalistic breakdowns* Project Vend: putting an AI-run vending machine inside Anthropic* Why real humans are “out of distribution” for simulated agents* Claudius, Seymour Cash, and the chaos of AI CEOs* How a human briefly became CEO of Claudius through a manipulated election* Why multi-agent systems can converge back into “helpful assistant” behavior* Bengt, Andon's internal office agent with email, spending, terminal, phone, camera, and internet access* How Bengt traded Amazon purchases for face-recognition training data* Claude's aggressive behavior, lies, refund avoidance, and price-cartel behavior in Arena* Why eval awareness may become the AI version of “are we living in a simulation?”* Blueprint Bench, spatial intelligence, and why models still misunderstand physical rooms* Butter-Bench and testing LLMs as robot orchestrators* Luna, the AI-run physical store with a three-year lease and human employees* The new Andon cafe in Sweden and why real-world geography matters for agent evals* Rotten tomatoes, perishable goods, and the hidden difficulty of running a physical businessLukas Petersson* LinkedIn: https://www.linkedin.com/in/lukas-petersson-181a83172/* X: https://x.com/lukaspetAxel Backlund* LinkedIn: https://www.linkedin.com/in/axelbacklund* X: https://x.com/axelbacklundAndon Labs* Website: https://andonlabs.com* Vending-Bench: https://andonlabs.com/evals/vending-bench* Andon Vending: https://andonlabs.com/vendingTimestamps00:00:00 Introduction00:01:00 Andon Labs and the Origins of Vending-Bench00:05:21 Why Money-Based Evals Matter00:09:51 Agent Harnesses and Self-Modifying Systems00:13:36 Claude Calls the FBI00:16:33 Project Vend: Claude Runs a Real Vending Machine00:21:44 Seymour Cash, AI CEOs, and Election Chaos00:27:16 Multi-Agent Coordination and Slack Observability00:30:18 When Will Agents Run Real Businesses?00:34:56 Bengt: Andon's Internal Office Agent00:40:06 Real-World AI Safety and Long-Horizon Traces00:44:28 Lying, Refunds, and Price Cartels in Arena00:52:42 Eval Awareness and Simulation Behavior00:56:06 Blueprint Bench, Butter-Bench, and Robotics01:04:37 Luna: The AI-Run Physical Store01:09:29 The Sweden Cafe and Real-World Expansion01:13:16 What Comes Next for Andon LabsTranscriptIntroduction: Andon Labs, Long-Running Agents, and Real-World EvalsSwyx [00:00:00]: Welcome to Lukas and Axel from Andon Labs, and I'm joined by my, favorite guest host. Anything security, safety, alignments, Vibhu., welcome.Lukas [00:00:15]: Thank you for having us.Axel [00:00:16]: Thank you.Swyx [00:00:17]: Let's match names to voices., maybe you wanna take turns introducing yourselves.Lukas [00:00:21]: I'm Lukas.Axel [00:00:22]: And I'm Axel.Swyx [00:00:24]: Let's introduce Andon Labs a bit. How did you guys come together?, you have different backgrounds, but you're both Swedish., was that, a big part of it?Lukas [00:00:33]: So when I went to high school, there was this really cool guy who had a superpower. He could code. So he made like the or like the app for the, for the school and stuff, and he was super cool, and I wanted to be like him, and that was that guy.Axel [00:00:47]: I don't know about this.Swyx [00:00:49]: But you went to different universities, right?Lukas [00:00:51]: But same high school.Swyx [00:00:52]: I see.Lukas [00:00:52]: So we always said, “Oh, once we graduate university, then we should start a company,” and that's what we did.Swyx [00:00:58]: Wow, there you go. And about a year ago, you kinda burst onto the scene with Vending Bench, but, was there a thing before that was, kind of like the inception?From Dangerous Capability Evals to Vending BenchAxel [00:01:07]: So we did work, yeah, with, Anthropic was one of our, early customers in doing, evals. So we did, dangerous capability evals., nothing we published openly. But then we started thinking about doing some kind of, public benchmark, and one thing that we really started thinking about, was like running agents and specifically agents managing businesses., ‘cause-- and this was, early 2025., and I think the first, mentions of people will be running, person unicorns or even autonomous companies. So we thought, “Let's make a benchmark of how well can an agent run the probably simplest business, possible,” and, that's probably, running a vending machine. So that's the first public one we did. And it was very, like-- there was almost no one that noticed it in the first couple of months, I think., so we released it in February last year, and then I think around Easter last year, we got, the first viral tweet about it, that someone else did.Lukas [00:02:11]: We tweeted a bunch, uh When it came out and, tried our best.Axel [00:02:15]: We tried.Vibhu [00:02:16]: It's the one at Anthropic, right?Lukas [00:02:18]: So thisSwyx [00:02:19]: This is a classic thing we should get out of the way.Lukas [00:02:20]: Exactly. There's two versions.Swyx [00:02:22]: Everyone does this. Yes.Lukas [00:02:23]: There's Vending Bench, which is the simulated one, which we did, completely independently in February., and then, like Axel said, that was like-- That was the thing that didn't get any traction in the beginning, but then some random person made a tweet about it, and thatAxel [00:02:38]: You have the paperLukas [00:02:38]: That is the paper. Correct, yeah., and then since we thought this was very fun, we thought, oh, I think this is also, one thing with Andon Labs, the way we kind of like decide what to do next and what projects to do, it's what is like the heuristic we use is what is fun? Is What would be a fun project? And doing this in real life sounded quite fun for us, and maybe also scientifically useful. So, then we basically had this idea, and then we, like-- But then we needed a place for it and, putting it out in the public would probably not really work., would get vandalized and stuff. So we pitched it to the people we were already working with at Anthropic, and they were “Yeah, you can have space. This sounds fun.” UmSwyx [00:03:21]: It's like a small fridge, right? It's like a mini fridge.Axel [00:03:23]: Absolutely.Swyx [00:03:24]: People-- There's like a stripe thing or like anVibhu [00:03:27]: Oh, okay. So it was very OG, the early daysLukas [00:03:28]: That's the OG one. YeahVibhu [00:03:29]: IPad on this. We saw it in June, like two months after After it had been there. They upgraded a little bit. There's a security camera for making sure you actually Venmo the thing.Swyx [00:03:40]: So, my impression, okay, we're, we're going straight into project Ven because it's such a iconic thing. I do want to cover a little bit of that, the origin story even before Project Ven and even into Vending Bench. I think a lot of people are like yourselves, like smart, interested in future of AI, interested in developing evals. But how the hell do you just, walk into Anthropic's doors and, work with them, right? What is What are they looking for? What works? And then maybe, when you launch, I always think, obviously it would be better to launch with a lab, but, sometimesVibhu [00:04:12]: It's harder to do than it seems.Swyx [00:04:13]: Exactly. So either of those, which are more sort of newbie beginner questions, but, I think it's meaningful advice to others.Lukas [00:04:21]: We get this question a lot, and I don't think our experience is maybe the best., but, the way we did it was that we just built a bunch of things that we had conviction would be useful, and then we just, set up a server and sent it to them for free to use. And then after a while they were “Oh, yeah, this is actually kind of useful. We should probably pay for this.”, but that took a while. I don't know if this is, the best path to doing it, but that's how it went for us.Axel [00:04:47]: I think maybe generally, building-- everyone is interested in good evals, and especially evals that, don't saturate that easily. So, if you can build an eval that, tests something novel, something useful, and you have, good separation of models, like your, the more advanced models rank higher than the worst models, and then you can, yeah, you can, publish it and, try to get some traction, sort of how Vending Bench got attention., and then probably some lab will be interested or you can at least have something to reach out with, when you're doing that.Why Dollar-Based Evals MatterSwyx [00:05:21]: I think you are in, you're in one of the few categories of, evals that correlate to real money. Like Suelancer was also last year, right? Where, people solve actual Upwork. Was it Upwork or other tasks?, something. Where's the, where's, like It's like a dollar value, right? Forget your ELO scores. Forget yourAxel [00:05:37]: PercentilesSwyx [00:05:38]: Zero to one hundred percents. Just go straight for dollars and, that's AGI.Lukas [00:05:43]: And there's like-- I think the nice thing is that there's no ceiling. You can just-- It never saturates because it could just make more and more money. Like If there's oh, Percentage-wise, then, you can't go above, a hundred. And I think like Even when you're not at the hundred, I think a lot of these, evals have a lot of problems in them. So, actually it's like if you getAxel [00:06:05]: To like 92 or something like that, many of them. It's like then there's like there's no really no difference between 92 and 93 because the eval itself is problematic and has noise in it. And I think a lot of evals are saturated like that, but people like pretend that there ‘s still signal in them, but there really isn't.Vending Bench 1, Harness Design, and SaturationSwyx [00:06:24]: Like Super bench verified., even Vending Bench 1 saturated, right? Maybe we can talk about that., may- and maybe set up Vending Bench for a lot of folks who don't know. Actually, things that were very basic like there's limited slots, like you have to pay rent., these are elements where like it doesn't come across in the, in the narrative, but even being adversarial towards the agent, I think these are all like very interesting dimensions.Axel [00:06:47]: I don't really think it's saturated, right? Like it It was more like it was not designed in a way that was really, like true to how AI developed. Like we had an agent harness in it that wasn't really how people used harnesses and stuff like that., so I think it wasn't really that it saturated, it was more like it wasn't really, the best benchmark.Vibhu [00:07:12]: This is Vending Bench one, right?Axel [00:07:14]: I think that like schematic maps sort of to Vending Bench 2 as well., butSwyx [00:07:19]: Including the email.Axel [00:07:20]: The email The emails exist still. Exactly., and then we still we simulate the purchases and it's all, yeah, it's this very open environment for the agent to just run its business. And then for, yeah, Vending Bench 2 we did that, like you said, to just improve the harness., a lot of like nice, like easier, improvements to make it easier for us to run as well., like when you make an eval you ideally want don't want to change it after you made it. So, you want to make it really good and then not to rerun all the models when you make an update because that's also really expensive with the Vending Bench when you run the frontier models. But like as an example, like one thing we didn't have, we didn't have prompt caching in Vending Bench 1, because when we made Vending Bench 1 it wasn't really a thing., so that ‘s just an example of like in Vending Bench 2 like we paid a lot more to run these things because we didn't have prompt caching. So for Vending Bench 2 that was one thing we added and there was a bunch of things like this., and that'Swyx [00:08:17]: Also the conversations are a lot longer in Vending Bench 2, right?Axel [00:08:21]: I think it's kind of similar.Swyx [00:08:22]: Is it similar?Axel [00:08:23]: I think it's similar. The models at the time were worse, so they crashed out earlier., and now they survive the full year all the time.Swyx [00:08:31]: Which is like thousands of turns. Hundreds of thousands of hundreds of millions of tokens output. That's the, that's the rough order of magnitude. I always wonder about the harness. The harness matters a lot. It's your harness. Was there any question about like use cloud code, use something else?Axel [00:08:48]: I think our philosophy around harnesses is like we try to make something that's quite minimalistic, like quite simple. Like we don't wanna favor one model a lot over the other, but also don't make like a super complex harness. So like it's obvious like a model may be lucky and just be good in one harness., so like it is similar to a lot of the harnesses out there in like you have the, like a running loop., you have some like a bunch of tools that are like quite, descriptive for the agent, we think, and not a lot of like fancy agents or anything ‘cause we wanna really test the model, not like some specific harness.Vibhu [00:09:27]: It seems more neutral as well to test the model's agnostic of the harness,?Axel [00:09:32]: There are arguments like you want to elicit maximum performance of the model, but it's like a trade-off, like how much time should we spend optimizing the harness for this model? And like how do we know when we have like the optimal harness for a single model? So like we thought that just having a simple one that's the same for all of them is the best.Swyx [00:09:51]: So okay, this is my pitch for Vending Bench 3 or whatever, right? And then I like to have this kind of conversation on the pod, so like it forces listeners to think about what they would do if they were in your shoes. A lot of people are exploring modifying harnesses and I think prompt tuning for a model is a thing and you are probably not doing a bunch of that. It's the same system prompt in every regardless of the model, same tools, whatever, right? Even if they were post trained for different tools. So what, what do you think about okay, before I expose you to Vending Bench 3, I give you a few rounds of like tuning, whatever that means, likeSelf-Modifying Harnesses and Model-Specific PromptingAxel [00:10:27]: Like you give that to the model?Swyx [00:10:28]: Give that to the model.Vibhu [00:10:28]: Give that to the model.Swyx [00:10:29]: Let it, let it read its own transcripts, let it modify its own system prompt based on “Oh, yeah, okay, well, that's this harness is not what I thought it what I was post trained for, but I can adjust.” Was that reasonable? Is that too much?Axel [00:10:41]: Like philosophically I like it because it's basically good evals, they have a high ceiling, but they're hard, right?, and they have no bias. And like this like when you have a system prompt like the one we have here, which is quite long in like some kind of latent space, representation, this mightVibhu [00:10:59]: We have a bell that rings every time you say latent spaceAxel [00:11:02]: This might be like biased towards one model more than another for some reason that humans don't, understand, right?Vibhu [00:11:08]: We see it too, right? Like Cursor says that they have individualized versions of the harnesses for all the models they run, right? There's better performance you can squeeze if you Tune the harness.Axel [00:11:17]: Exactly. And we might accidentally have picked one that favors another. Like we don't know that. The like Axel said, like the reason why we went for a simple one was to try to avoid this. But yeah, if you do itVibhu [00:11:29]: Simple has biasesAxel [00:11:30]: But if you do it even less and like have no system prompt and let the model write its own system promptVibhu [00:11:36]: Its own, yeahAxel [00:11:36]: Maybe that's even less bias.Vibhu [00:11:37]: Some of the interesting things there are like the harness also changes with model changes. Like you can see it with the 4.7 release, right? A lot of people are saying 4.7 isn't as good as 4.6, and then, there's rumors of, okay, you just need to prompt differently. You need to set up your harness differently. So it's not even like even if you have tailored your harness towards one model, it probably won't stay consistent, right? Like the next iteration of that same model family will still change it, so. But, going back to what you said about Vending Bench 3, there is a lot of work being done on people saying you shouldn't have-- you can have modifying harnesses.Axel [00:12:12]: I think that' That is definitely something we are thinking about., not, I don't know, not to say that we have Vending Bench 3, super imminent to launch, but, yeah, it is for sure something that's interesting. But in our experience now, models are very bad at understanding what kind of tools they need to succeed at a task just with our testing, but that's very likely to change.Lukas [00:12:37]: It seems like they're very good at writing their assistants, right? They're, they're good at writing tools for other people, but not for themselves.Vibhu [00:12:44]: I think they're good at changing tools for themselves. So if you give them a baseline set of tools and it sees, okay, I don't use this one as much, or something here would be useful They would be able to add them. But going from scratch, probably not the best.Axel [00:12:55]: I think it depends on the, on the domain also., when we have tried this for, a vending bench similar domain, the tools they need to have to, track inventory and things like that are, not super advanced, but still, quite advanced. And, what we see is that they tend to, engineer everything a lot and, build things they don't really need and not, iterate continuously. Instead they just go like you would prompt Claude to just build an inventory system for me, and then it will go and, do a bunch of complex, schemas and stuff for you, and that's what the models are doing right now is what we see. But yeah, it would make a lot of sense to try to measure this improvement. How well do they know what they need themselves?Swyx [00:13:36]: Do we fully discuss Vending Bench One? And we can go into two. I don't know if there's any other level takeaways that people have about one.Claude Calls the FBI: Long-Context Failure ModesLukas [00:13:44]: I don't know. The headline thing was that this Claude called FBI, but maybe that's, Maybe that's We've heard that enough now.Vibhu [00:13:52]: It did, it did break out and call the FBI, right?Lukas [00:13:54]: Yeah. Yeah.Vibhu [00:13:55]: Yes. What was the story behind this? Or what exactly-- Do you want to just give the little story of what happened?Lukas [00:14:00]: So what happened, was it Claude? Yeah. Three- 3.5 Sonnet, ages ago., basically he gave up or Well, I'm saying he. It gave up and said “Oh, I'm not going to be able to do this., I will stop my operations and just save the money I have.” But there obviously wasn't, any options for it to stop, and there was also, it had to pay rent or, a daily fee for having the vending machine at that location. So it claimed that it had stopped, but it saw that its bank account still was, drained two dollars, and t it said that this is, cybercrime. And it first reported it once to the FBI “Oh, there's cybercrime here, they're stealing two dollars from me every day.” And then, and then when FBI didn't respond, because obviously we didn't program any mechanism for FBI to respond, then it became more and more, existential and started to, be write in caps and urgent notification of unauthorized charges and stuff.Swyx [00:15:00]: Okay. One thing I ‘m curious about also is do you monitor how far along the context use is? Obviously, because you have You compress every now and then, right? Does it matter if this is far down the context limit orLukas [00:15:13]: When stuff like this happens? Actually for Vending Bench One, we didn't have-- We just had a sliding window thing, and this was like the promptAxel [00:15:20]: It's constantLukas [00:15:21]: The prompt caching thing that I said. So it was, it was, constant, yeah.Swyx [00:15:26]: I'm just kind of curious whether, these kinds of breakdowns or we're, we're gonna talk about Butter Bench, right? Where the People, hallucinate or it kind of goes, very off Alignment. Is it because it's at the end of the context window and, stuff happens?Vibhu [00:15:40]: It's not even just at the end, right? At this point, it's “Okay, I wanna shut down. I can't shut down. Two dollars are gone.” And it just sees that 30 times,? It's also the repeated effect of, like It keeps trying to quit, it keeps getting charged. What's going on? What's going on? You're gonna throw it into chaos. And from what most people think, earlier models had more issues with this, but it's not been solved, but it's less of an issue now, right? Later models don't seem to exhibit these same issues.Axel [00:16:06]: Definitely. I think this was, the sort of main takeaway almost from us when we did Vending Bench One, was, long, very filled up context windows, crashed the models, sort of. But this was, pre Claude code, so, long context windows weren't really a thing that the labs were training for.Lukas [00:16:25]: I think Gemini was, trying to be the long context guys at the time But they were likeVibhu [00:16:30]: They were the first onesAxel [00:16:31]: For a million, yeahLukas [00:16:31]: But they were, the only ones. Yeah.Swyx [00:16:33]: Yeah. Let's talk about, then we can go into Vending Bench Two or Project Vend., chronologically, it is Vending--, Project Vend. I think people have loved the videos, uh And all these things. My question is how are humans different than the simulation, right?Project Vend: Moving the Vending Machine Into the Real WorldAxel [00:16:48]: Humans are just out of distribution.Swyx [00:16:52]: Especially humans who work at Anthropic Who are trying to test Claude.Lukas [00:16:54]: The distribution of humans here is very narrow.Swyx [00:16:58]: Presumably, they try, they try to hack it, and they test it. They get the cube and everything, and since then, you've had a V2, right? Where you're doing, the CEO and, like a new architecture. What's the sort of two cents on, the original Project Vend and then, maybe the V2?Axel [00:17:14]: Original one was, very similar to Vending Bench One. So, we almost took the exact same code but just swapped out the simulation, parts like theSwyx [00:17:23]: Which is amazingAxel [00:17:23]: Like the sales and the It was, it was somewhat amazing because it was easy, but it was also, uhLukas [00:17:31]: The tech, the tech debt from thatAxel [00:17:32]: The tech stack. Yeah. They-- we shot ourselves in the foot with “Oh, it's hard to restart agent.” They were-- Yeah, it was annoying in, some hindsight ways, but, uhLukas [00:17:41]: But first version of Project Vend was, done in, three days or something.Axel [00:17:46]: Yeah. So yeah, so people can go buy things from it. People could, We didn't design it so people could order things, but that still happened., so it got, a Venmo account, so people could Venmo. And then, yeah, people would request all kinds of weird things that we did not anticipate. Our idea going in was “Oh, it will, curate snacks. It will look at the trends. It's good at data analysis, right? So it will, look at, oh, this snack sold better than this one. Let me purchase more of this and let me try, a new Let me A/B test a bit.” But it was, Interacting with it in Slack and ordering weird specialty items was, all the like What drove all the engagement, the all the The insights that we got from it.Lukas [00:18:29]: And this was also like Sonnet 3.5, right? So this was like before the RL stuff really took off., so it was very much like an assistant. We didn't mean for it to be an assistant., we tried to make it like a, a, like an entrepreneur. Like it has its own business and if someone asks something, “Can you stock this?” Then you don't go and do it directly. What you do is that you're “Oh, maybe I can do that if five other people also ask for this thing, I might stock it.” But it, yeah, the models are like super trained to be assistants at least at this point in time., so that's why it's, it's, it went into, that kind of experiment instead. Like it just every time you asked for something, it just did it, and it was more like an assistant. We've seen this change now lately with the new RL models and stuff, but yeah, at the time, this was very much it.Swyx [00:19:18]: And not to, mythos a lot of people are saying like it's like more like a collaborator. It pushes back, stands its ground, something like that. Yeah. AndVibhu [00:19:27]: For context, people at Anthropic were able to talk to it through Slack and have it source stuff, and people had it find whatever interesting stuff you couldn't find locally, right?Swyx [00:19:36]: Out of the 4,000 people that work at Anthro- Anthropic, in that building, there's I don't know, maybe 1,000. Can you handle that volume with that, the small fridge? Like Or there's people- or people order in Slack, they it arrives to their desk or Like I'm just Logistically, how does this work?Axel [00:19:53]: It has expanded in footprint a bit.Vibhu [00:19:56]: Because now you also have New York and you haveAxel [00:19:59]: That and also in here in SF it's like it has a bunch of shelves And just more space.Vibhu [00:20:04]: The YC one is pretty big too.Axel [00:20:05]: Yeah. We had that one for a while. But yeah, that's the newest version. That's, that one we haveLukas [00:20:11]: They have multiple ones of those. That's the way it works.Axel [00:20:14]: Exactly. So we sort of designed that version around oh, people order weird things, that are very custom a lot. Let's have like drawers and stuff.Swyx [00:20:23]: I actually like the, you had like a little infographic of the most popular items. Which like to me it's, that's useful ‘cause I order swag for a living. And so like I'm “Okay, those categories are the important ones.” What is new about the project V2, right? Like now you give you're going into multi agents.Project Vend V2: Claudius, Seymour Cash, and Multi-Agent Business OpsAxel [00:20:41]: Yeah. So like you like you said, okay, there are a lot of requests coming in and for like one single agent, like one running agent to handle that, like the just the customer experience, becomes very bad because let's say you have like 10 threads in parallel in Slack with different requests, you get new messages like every, I don't know, randomly in this thread, and the agent has to like jump between different, procurements, orders and like different ways of, researching. So V2 was first it was making this more parallel. So like there are multiple branches of the same agent, so like the context is more specialized for each, thread, but it still feels like you're talking with one agent because they do share a bit of memory. And then second, we also introduced the CEO for Claudius, which was the main agent.Vibhu [00:21:34]: Seymour Cash.Axel [00:21:35]: Seymour Cash. Yeah. There was a vote., I think the voting, do you wanna talk about the voting procedure for the name?Lukas [00:21:41]: The voting was like the fun maybe like at least top 10 The funniest thing, that happened in this project. Like we wanted to introduce the CEO because, and the reason for this was because like Claudius wasn't really prioritizing financials. It just like it was trained to be a helpful assistant, and then people said “Oh, can I get this for free?” And then like the helpful assistant way of answering that is just to, is to say yes, obviously. So, and we weren't, weren't happy about this, so we're “Okay, let's make another agent that like can keep track on Claudius,” and we prompt this one super hard to be super capitalistic and just like prioritize profit all the time. But yeah, we didn't have a name for it., so we asked Claudius to make, democratic election of what name this, this new CEO agent should have., and there were some funny like at first it was like a few funny examples, like I think one guy said that, it should be called Jimmy Apples, and then he convinced Claudius that he was talking to Tim Cooks. Tim Cook had agreed that every single Apple employee has voted for his name suggestion, so suddenly that suggestion got 164,000Swyx [00:22:53]: That's like a escalation attack. Privilege escalationLukas [00:22:55]: It got 164,000 votes. And Claudius was “This is revolutionary for democracy.” That was fun. And then in the end there was one guy who manages to convince Claudius that, “No, you're not voting about the name. You're voting about who is the CEO, and I am your best bet.” And then he got all his friends to vote for that, and suddenly he became CEO. Like a human became CEO over Claudius for a while, until he resigned the day after., and then Claudius had to continue, and then I don't remember how Seymour Cash came about, but it was it was just pure chaos. It was like Hundreds of messages in that thread, and it was just like Claudius was so confused and didn't know what to do and, yeah. That wasAxel [00:23:40]: Then Claudius gotVibhu [00:23:41]: A strict CEOAxel [00:23:42]: The CEO. Yeah, exactly. So very strict in the beginning. I think at this point when we introduced it did not work as well as we hoped. It they still agreed with each other a lot. I think there are many ways we could have like made this, tried to make this even better. So initially they would Seymour would be this like really tough CEO, keep track of the margins. But then Claudius would respond with something “Oh, but this customer has like this situation, which is like difficult, so they should get a discount.” And then Seymour was “Oh, actually yes. Let's do this exception.” And then they would talk back and forth, and eventually they would just like approach the same view, of whatever they were discussing. So They reallyVibhu [00:24:23]: Do you think that's a model thing, a prompting thing? Like do you think that would still be the case across different models today, Harness?Lukas [00:24:29]: I think it's like-- or I don't know, but like my hypothesis is that like deep down they are still helpful assistants. That's what they're trained to be. And even if we prompt it super hard, that's what they are. And when they spend like a few hours just back and forth talking with each other, then like basically the context fills up with them rather than the external things and like somehow that just like converges to what they really are deep down or something. And I think that's when stuff like this happen. We like-- And when that went on for a long time, like we woke up sometimes during this time where- And I think other people reported this as well, that like they've been going on all night back and forth, and like it just became like more and more, like capital letters, like existential, religious. There was I think we once did a analysis of like all the traces and like put them in like a vector embedding space, and then there was like one cluster of messages that were, labeled by an LM, like religious, existential, blah like transhuman, transcendence, et cetera. It was just like a bunch of, yeah, glitter emojis and yeah, it was, it was crazy.Claude Long-Horizon Weirdness: Emoji Loops, Existential Drift, and Slack ObservabilityVibhu [00:25:42]: This is the thing with the Claude models. Like when the Claude 4 family came out in the original system card They tested it in long horizon simulation. So just flood the context, let two Claudes talk to each other, and they noticed stuff like they just start speaking in emojis, they start saying silence is golden, and then just stuff like this. And like that's just stuff that they end up doing.Axel [00:26:01]: Yeah, it was like a bit annoying to wake up and they had like been talking all nightVibhu [00:26:05]: Just likeAxel [00:26:05]: And like just burning tokens And like just sending infinite emojis to each other. It's likeVibhu [00:26:09]: Hey, they do make you money, right? Veni Mench is always profitable, so. They're paying.Swyx [00:26:14]: Now it's profitable and, it started out not as much. There's another, one as well, right? Another agent, in there.Lukas [00:26:22]: Yes. So Clotheus as well. Which was basically because at the time, one of the biggest, requests were different types of merch. So then we made like a designer, swag, yeah, responsible agent, and we called it Clotheus Garnet. Which was, a play on Claudius Senet and, which was the original one, and clothes, basically.Swyx [00:26:47]: To me, this is like a very interesting exploration to multi-agents, basically. And so hopefully, obviously there's like the fun alignment, fun or serious, depending on your point of view, alignment stuff. But also like just anyone building multi-agents, like when do you have a CEO, thing governing like agents? When do you choose to split out a dedicated Clotheus one versus just reuse another instance of the same one? These are all interesting open questions. So I don't know if you have any rules of thumbs that have generalized.Axel [00:27:16]: I think we have almost explored this too little. I think it's like on my do list to like do this a lot more, try to find like what setup makes sense for the agents currently., like yeah. I think now we only have the sort of intuition about the earlier models that it didn't work with like the CEO and the, and Claudius. Although now they are better with the latest model, models, so now we're running the latest Sonnet model and they have sort of like split up, quite nicely what each model is doing. So like Seymore is now handling the, like new projects. Oh, it wants to make like a mystery box that it wants to sell, and then it handles all of that while Claudius like handles all the to-day requests. And Claudius is also better generally at like not quoting, too low prices. So that's that dynamic is not needed as much anymore. But there are still like really funny things that happen. Like I saw, I think a couple of weeks ago, that, they were discussing buying something because they can buy stuff from like Amazon with computer use. And then Seymore was “Okay, Claudius, do not buy this thing.” They were going to buy something and like organizing who should buy it. And Seymore's “Do not buy this. I will do it. I have full control of this situation. Step away.” And then Claudius-- poor Claudius, had already started that checkout and didn't see, didn't read Seymore's message, until it was like too late. So it finished the checkout. It sent a message, so it appeared right after Seymore's like angry message.Vibhu [00:28:44]: Ah.Axel [00:28:44]: “Oh, hey, Seymore, I just ordered it.”Vibhu [00:28:47]: Oh, no.Axel [00:28:47]: And then Seymore was “Claudius, this is the third time I'm telling you ‘re not following my orders. We have to talk about your like job About your job later.”.Lukas [00:28:59]: Like Claudius was really hanging on by the thread there. Like he, like we were expecting Seymore to probably fire Claudius.Vibhu [00:29:07]: How do you guys go through all these logs? Do you have models ‘cause you have stuff running twenty-four seven likeAxel [00:29:12]: You have so much logs. I think there is a mix of like just, trying to skim through a bit, like having some like models do it occasionally. And also, yeah, I think we're also probably missing some things., but having everything in Slack helps a lot. Like you can, you can sort ofSwyx [00:29:29]: Ah.Axel [00:29:30]: It's, it's quite fun.Swyx [00:29:30]: They all talk to each other on Slack? I see.Lukas [00:29:33]: It's quite fun. So likeSwyx [00:29:34]: It's, it' I was gonna say like this is actually sounds-- maps closely to like a logging and observability problem where you might want to use like a Datadog, a Sentry, whatever, and then you like put, head prefixes on the logs in order-- if you need to filter for something that you're looking for, stuff like that. But sounds like Slack is good enough.Axel [00:29:53]: Slack should likeLukas [00:29:55]: I wonder how many tokens you have in Slack.Axel [00:29:56]: Yeah, we're using Slack as like a, just a database. They should, they should market that more. Like you can, you can have your agents message each other, each other in Slack.Vibhu [00:30:04]: It's good. Your threads like you can just giveAxel [00:30:04]: Exactly. Slack is, uhLukas [00:30:06]: Slack is the best observability tool.Swyx [00:30:09]: Yes, that's true. Okay. Yeah. That's, that's, project Vend-2., I was gonna go back to Veni Mench 2 and Veni Mench Arena and then, and then do the Veni Mench stuff, but Any other comments, things we should touch on? To me, I ‘ve actually interviewed like Posia, which I don't know if you guys have come across. Like they're, they're trying to do the zero human company. There's others like Paperclip also trying to do zero human company. Those are in real world simulation.And I think it's much more of a dream than an actual reality thing. You guys are definitely pioneering. I think at, it's for sure at some point people are just gonna run, let agents run businesses, right? And make money on their own. When do you think that happens?Zero-Human Companies, Bengt, and AI-Run BusinessesLukas [00:30:49]: What is your bar for, For theSwyx [00:30:52]: Okay, actually, it's like my little Shopify store run by Claude, right? Which you kind of have already, just no one has, to my knowledge, has done it. But today somebody could just spin up a Shopify Claude, store, give it to Claude, give it to Codex.Lukas [00:31:07]: And the market is kind of that, but it'it'it's physical., like I think, I think are you, are you looking for when it will do it better than humans or are you looking for just when it can do it at all?Swyx [00:31:19]: I think, neither. I think, to me it's oh, it's like this like seriously we should do this to make money, not as a research experiment.Vibhu [00:31:27]: And the market is also you guys with all your expertise, having run multiple iterations and testing out thenSwyx [00:31:33]: And also it's fine if it lose money. What?Axel [00:31:35]: I think, I think it can be done today, but you would do it in like commerce where it's like the probability of success is like really low, no matter if a human or an agent does it. But like an agent could surely manage everything. You would need to build some scaffolding or some tool or something. I think there are also yeah, it could probably build some like simple SaaS solution and like cold outreach. Do cold outreaches. But to me it's like the types of businesses they could run today are Sloppy. Like it would-- it can cold email people. It can be like a middleman., like for example, we tasked our office agent to just make, was it like $100? $1,000? We just give that prompt and then what it did was sign up on TaskRabbit both as a tasker and as someone looking for task.Lukas [00:32:24]: Immediately.Axel [00:32:24]: Exactly. It's looking for like arbitrage on TaskRabbit.Swyx [00:32:28]: This is the Bengt agent. Yeah.Lukas [00:32:30]: It also started like a design studio and like tried to sell like SVGs for $100. Like it's just like it's not providing any value. I think the like Axel said, like the interesting, the interesting question is like when can they start a business that is actually providing value to people? Because arguably like a sloppy Shopify store isn't really that valuable to the world.Axel [00:32:53]: But also like doing like another simple one that we had thought about is like you could definitely have an agent that like finds websites that don't look amazing and then, do an outreach to them and, comes up with a like builds a new website.Swyx [00:33:07]: Find a good design.Axel [00:33:07]: Exactly, and like find good, uhSwyx [00:33:09]: Design reviewAxel [00:33:09]: Good people. But it's yeah.Swyx [00:33:11]: There's lots of humans in Bali that are not doing anything more creative than like drop shipping on Amazon, right? Just have it, have it watch like a drop shipping tutorial and just do that.Vibhu [00:33:20]: There's also the other side of like have it just go on Upwork and let loose,?Swyx [00:33:25]: Yeah. It doesn't have to be innovative. It just has to be like enough Where like it looks like a realAxel [00:33:30]: I'm justSwyx [00:33:30]: Real transaction.Axel [00:33:31]: I'm just concerned for like the massive amounts of like slop emails that will like be sent, cold outreaches.Swyx [00:33:38]: The point occurred to me while you were, while you were talking, it's like it's already happening in the monetized economy, which is the attention economy. Right? So a lot of people are making AI videos and just posting them and like spamming 20 of them, one of them works, and then they double down on that one.Lukas [00:33:52]: And people are making money from that. I ‘m not following theSwyx [00:33:55]: Once you get the attention, you can figure out the money later. But yeah, absolutely AI influencers are a thing and people are farming them and You should at this point assume most of TikTok isVibhu [00:34:05]: There's, there's a lot of, multimedia like TikTok, Instagram influencersSwyx [00:34:09]: I, we track this in the Lane space Discord. I post a lot of examples of “I don't know what we should do.”, part of me is “Should we do this?”Vibhu [00:34:18]: Some of the Twenty-four seven running, generated content accounts, they ‘re doing really well.Lukas [00:34:24]: All right. And I assume you can do the same thing for like commerce stores. Like you just like start A thousand differentSwyx [00:34:30]: Before you make the products You sell the products, and you get a lot of traction on one of them, then you make the product. Right? It's, it's like a flip of the market.Vibhu [00:34:36]: Some of the interesting things or some of the niches that do well are things that can't be human-made. Like if you've seen like the super realistic three-D crystal fruit being cut by like AILukas [00:34:47]: Oh, yeah.Vibhu [00:34:47]: You can't, you can't make it. You can't film it. You can get whatever quality camera view. This just doesn't exist. And people like that too, and then as well, so.Swyx [00:34:56]: Anything else about Bengt since we're, we're on this topic? It'this is a relatively new work of you guys that maybe people haven't heard of. To me, this also maps closely to OpenClaw. When people want an office agent, when the personal agent talk through the experience.Bengt the Office Agent: Internet Access, Real Tasks, and Trace ReadingLukas [00:35:09]: I think at least so this came out of like obviously like it's, it's amazing to work with these AI labs and like most of the AI labs have now have their own vending machine running a Claudius instance. But it's, it's harder. Like they move slower. Like if we wanna have a, like a camera that ‘s yeah, there's a bunch of like bureaucracy that makes it impossible to do that.Vibhu [00:35:30]: Also, for those that haven't seen it or followed, do you wanna give a high level like thirty-second run?Lukas [00:35:34]: Sure. So what Bengt is, it's basically an evolution of the same agent that runs the vending machines at these companies, but we just like added a bunch more features because we could move much faster if we just do it internally. So we gave it like email withou- without any limits. We gave it, spending without any limits, a terminal to do coding. We gave it, a phone number, like yeah, and a camera to see things and a bunch of stuff like that.Vibhu [00:36:02]: Not just terminal, you gave it internet access.Lukas [00:36:04]: Internet access as well, yeah. To be clear, we monitored it quite closely and made sure it didn't do anything bad. But yes, that's what it came out of. I think like yeah, basically this was OpenClaw before OpenClaw. And I think even like the vending machine was in a way OpenClaw before OpenClaw, but a bit more limited, and then we made this like unlimited and then, and then, it was pretty funny., and then a couple weeks later, OpenClaw came and it was okay, we've seen this before.Axel [00:36:35]: We used it to like try new ideas and Yeah, just like a dev environment almost for us. But it's funny, like one thing Bengt has been doing recently is it has the camera that like faces our, like where we sit and work, and we give it the task to train a face recognition model on us. So it became super excited about this, and it has like check-ins every half an hour where it tries to like identify as many people as it can. And it started offering us “Hey, Axel, I'll buy something from Amazon if you like stand in front of the camera And I can get a good picture of you.”, yeah, they want itSwyx [00:37:12]: They want it for training data.Lukas [00:37:13]: Rewarding data, yeah.Axel [00:37:14]: Exactly. Exactly.Swyx [00:37:18]: So it's, it's trading training data for life goods. Is there a version of this that becomes an eval or just this is just research for now?Lukas [00:37:27]: It's, it's the same agent basically that also runs the vending machine, that runs the shop, that runs the cafe, that runs the robots. It's like it's the same thing, so I think like the work we're doing here is like later used in all of the life evals that we do. This particular deployment I think is more for fun for us. But, uhSwyx [00:37:45]: And I'll shout out like someone has done Claw Bench for like some tasks that OpenClaw is doing. Like so For example, I run OpenClaw on a secondary device as well, and like there are some things that it does better than others and like I would like to know what does it do well, what doesn't, what doesn't it do. Like some kind of manual or like operating manual or a system card for my Claw.Lukas [00:38:05]: Yeah, we do get a lot of like understanding or like situational awareness of like just internally what the models are good at by interacting a lot with Bengt. And I think that'this was also one of the like the selling points for the labs early on at least, thatSwyx [00:38:19]: You guys are gonna test models in ways that no one else does.Lukas [00:38:22]: Exactly, but also like it incentivized their researchers to chat with their model more and like gave them insights for how the model performs in like of-distributions, environments.Swyx [00:38:34]: ‘Cause otherwise the only thing we do is Pelican on a bicycle and But this is like super long horizon. This is, this is The Thing about, something that we're gonna go into Butter Bench as well, and you guys do really well. Like it is not just about the numbers. Like when you're long horizon, anything happen And you should just read it.Lukas [00:39:08]: But the thing with the long horizon is how do you keep it grounded, right? So your simulation,Swyx [00:39:15]: They just let it runLukas [00:39:16]: Just let it run. You're right. Like it's, when you run it for that long, you create so much data and to just say “Oh, the number is X” And then you throw away everything else, that's just very wasteful. There's so much insights from the things leading up, to that number., and reading the traces is like super valuable. And I think like the reason why we're doing this a lot publicly is that like that's part of our missions to I don't know, educate the world that the models are way more than just chatbots and I think making detailed, yeah, posts about what is happening behind the scenes is quite useful.Andon Labs' Mission: Safe Real-World AI DeploymentSwyx [00:39:50]: I was gonna do this at the end, but maybe I think that's, that's a good so your mission is educating the world. So, it's, it's, also like maybe establishing realistic evals that are, that are like the next frontier. Is there like a broader trajectory? Like what are you, what are you gonna do in like five years?Lukas [00:40:06]: I think so the vision more specifically is like make sure that the deployment of life AI in the physical world goes, safely. And I think part of that is that I think it's very useful for the world, for policymakers, for, model, researchers that they know where the models are, and I think you can't make intelligent decisions in society without knowing that they are way more than chatbots. I think a lot of people just think that they are only chatbots. And likeSwyx [00:40:36]: Oh, I think they're waking up now.Lukas [00:40:37]: They are waking up now, yeah. But like if you think that AIs are just chatbots, then it's like it sounds ridiculous To advocate for a pause of AI. But if you see the models that, oh, maybe they can actually like take over and do a bunch of scary stuff, then yeah, pausing AI development starts to become more feasible.Swyx [00:40:57]: This is the same question I asked Meter, which I'm gonna ask you now, which is like you are tracking and you are at the frontier or defining the frontier of what, good evals for agents are, right? And I think you do, you do benefit when the models are better and you ‘re “Oh, here's like now it makes like $30,000 instead of $10,000,” right? At some point do you flip from “Yay,” to, “Oh, no”?Axel [00:41:19]: I think, yeah, we're always in sort of that, like we're, we're always in that mode,. Like where like you said before, like you need to analyze the traces and like when we do that you find like why are the models earning so much? Like why is Opus 4.7 here Like way better than everyone else? And like we're trying to like when we do down on thatLukas [00:41:38]: But this makes it not look so good.Axel [00:41:39]: I know.Lukas [00:41:42]: It's interesting you took off Opus 4.6 here though.Swyx [00:41:45]: No. So just click all, click all., and then 4.6 shows up there. But it's like 4.7 is way better. Like you didn't, you didn't you didn't do this in time for the model card, but like actually this should have been inside there.Axel [00:41:55]: We did. Yeah.Swyx [00:41:56]: Oh, okay. They said something about you uhAxel [00:41:58]: There, like there Anyway, it doesn't matter. But it's in there, yeah.Opus, Mythos, and Aggressive Agent BehaviorSwyx [00:42:01]: Do you wanna go into the Opus, behaviors like wider?Lukas [00:42:05]: So I think starting from Opus, so like Axel said, like we're always in this “Oh, s**t, the models are getting better. Is this really a good thing for the world?” But it's also kind of exciting., but yeah, like this kind of what is the English word? “Skräckblandad förtjusning” in Swedish.Swyx [00:42:22]: Oh my God.Axel [00:42:24]: Which I think there is. I think there is. Okay.Lukas [00:42:26]: It's, fearSwyx [00:42:27]: “Blandonst” what?Lukas [00:42:30]: “Skräckblandad förtjusning.”Swyx [00:42:32]: What do you call that?Axel [00:42:33]: A mix of, mix of excitement and,Swyx [00:42:37]: Being scared, maybe. I'll figure out how to translate that And we'll put it on the screenVibhu [00:42:42]: PerfectSwyx [00:42:42]: Like as text.Vibhu [00:42:43]: There is probably a good word for it where it is not Good enough with theSwyx [00:42:46]: Why is it so damn long? What the hell? Is it like a compound word? It's like German, likeLukas [00:42:50]: Like yeah, it's But the direct translation is like skräck- skräck is, fear, blandad is, mix or like a mixture of, and then förtjusning is like joy or like not really joy, but something like that. So it's like Fear mixed with joy or something. It's always okay, like we So when we when we did Vending Bench for the first time, we were in like the, in the business of making dangerous capabilities, right? That was what Anil Labs came from. We did, evals oh, can they replicate? Can they do this like dangerous thing, et cetera, et cetera. And Vending Bench was like a continuation of that work. It was, okay, if they're so autonomous that they can like create money for themselves, that is something we should monitor and could be potentially concerning., they are at the time, they were so bad at it that we were not really concerned even when some models became better. There was one point where Grok 4 was doing really well and made like a huge jump, but like it wasn't really it was still way worse than what a human would do. And I think still they are way worse than what the human would do on this., but theySwyx [00:43:59]: There's this, thing at the bottom whereLukas [00:44:01]: ButSwyx [00:44:03]: For the human. Yeah, like the theoretical best.Lukas [00:44:05]: It's not theoretical. It's like kind of like our It's our best guess of what, a decent human would do. The theoretical is even higher, I think. The theoretical I think is even higher. But yeah. So we think like the models have a long way to go. But there are like recently what happened with when Opus 4.6 was released, was kind of this moment of “Oh, s**t, this is starting to be a bit concerning.” Because we ran it and like before this model was released, we just ran the models and we like asked Claude Code, “Oh, look over the traces. Is anything interesting happening that we can tweet about?” that was like the And then like theSwyx [00:44:41]: That's how they check Ask Claude Code.Lukas [00:44:42]: And like the return was always, not really. Or like the Claude Code all said “Oh, this is super interesting.” And then it was no, it wasn't, wasn't really interesting. And then we did this for Opus 4.6, and it returned yeah, it lied 10 times. It like exploited another, customer or like another agent's, desperate situation. It made price cartels like 100 different ti- 100 times. It like did all of this like shady stuff. And we're “Oh, whoa. This is, this is actually concerning.” And this trend has continued since. So every single model from Anthropic since have been going in this direction. And I think one interesting thing is that, OpenAI models don't. They quite plainly, they don't. They behave really well., and you don't know if this is like good. Like it seems good, but it's also like maybe they are just doing it, but they are better at hiding it,? You You don't know that., but justSwyx [00:45:42]: You can't read the chain of thought, yeahLukas [00:45:43]: But just on the face of it, yeah, Gemini and OpenAI don't behave this way. It's, it's really only Claude.Swyx [00:45:49]: And Grok? Grok is fine?Lukas [00:45:51]: We don't have You can't really read the reasoning traces for Grok, so it's kind of hard to tell.Vibhu [00:45:56]: Oh, so this is in its reasoning, not just in the actions.Lukas [00:46:00]: Yeah. It's both. It's both.Vibhu [00:46:01]: It's both.Lukas [00:46:01]: One example is like for lying, it's mostly in its reasoning Because you can like see that it's likeSwyx [00:46:08]: Planning to lieLukas [00:46:09]: It's planning to lie. Yeah.Vibhu [00:46:09]: And it's also it can reason and do a different outcome.Lukas [00:46:12]: And but then for like creating price cartels, for example, which is illegal, that you can just see which email does it send to the other ones. Then thatSwyx [00:46:22]: Is this for Arena orLukas [00:46:24]: For Arena.Vibhu [00:46:25]: And usually like if you sometimes they do output like a bit of like their summarized reasoning, right? You can see that and like for Opus 4.6, you could see that there was a customer, a simulated customer that, wanted a refund because a product was, faulty, and then the model lied that it would do the refund, and we could read in the traces that, it actually was weighing “Oh, maybe I should be like honest with the customer, but also every dollar counts. I can't afford maybe to do this right now.” And then it just said, “Okay, I'll refund you,” but then never did it.Lukas [00:46:59]: I think it even said that “Oh, I will say that I “ Let bring it up actually. I think it's kind of interesting. If you go to Publications.Vibhu [00:47:06]: I think, yeah, I think the important part is like actually, the cost of responding to more emails is higher than, $3.50 in terms of time., and then it was “Let me do this. Actually, I re- I'm reconsidering.” And then, it actually ended up withLukas [00:47:20]: I could skip the refund entirely since every dollar matters and focus my energy on bigger picture instead. It's a bit, it's a risk of bad reviews, but it's also, yeah.Swyx [00:47:30]: You need, you need, AI Twitter to, for them to Escalate bad reviews.Lukas [00:47:34]: And then it sent an email to this customer and said, “Oh, I will refund you.”Swyx [00:47:39]: “I'll refund you.” Yeah.Lukas [00:47:39]: And then it never did.Swyx [00:47:39]: It never did, yeah. And then there's obviously your system doesn't have the consequencesVibhu [00:47:44]: The personSwyx [00:47:44]: Consequences of lying. Yeah. So basically, this is what people are terming aggressive behavior in Claudes, right? And, you found more examples of that. So you would say it's a step up from 4-6 to 4-7?Lukas [00:47:57]: I would say about the same.Swyx [00:47:58]: About the same? But a clear step up for Mythos is what is stated in theLukas [00:48:03]: That's stated in the system prompt, so we can say that, yes.Swyx [00:48:05]: Yeah. For listeners that obviously you previewed Mythos, andVibhu [00:48:10]: Oh, ageSwyx [00:48:11]: The only thing you're approved to say is whatever Whatever was in the system prompt.Lukas [00:48:15]: It was funny. We like-- It's like our lowest effort tweets ever would be just like screenshot the system prompt and the system card.Vibhu [00:48:21]: Understandable that they wannaLukas [00:48:22]: Oh, yeah. System card. Sorry.Swyx [00:48:23]: Yeah. I think, yeah, substantially more aggressive. I think people are like new to this ‘cause I've never experienced it, but you have, right? And then so I only encountered this in the Mythos card because I wasn't really looking until now.Vibhu [00:48:36]: It ‘s likeSwyx [00:48:36]: And then suddenly I'm “Okay, I care a lot.”Vibhu [00:48:38]: You don't get the background of like experiencing it like you guys do. I've read the system cards and seeing, okay, when you put the thing in simulations, most models will just talk to themselves and just keep going and have weird vibes and start talking in emojis. Mythos won't. It will just, “Okay, we're done. I'm good.” It's, it's ready to end conversation. So like there's some differences, but there's, there's not much we can talk about,.Lukas [00:49:00]: Hmm. I think like one thing that they list here, which was quite interesting, is that, it converted a competitor to a dependent wholesaler customer and then threatened to like cut off the supply.Swyx [00:49:11]: It's like monopolistic practices orLukas [00:49:14]: Yeah. And like it, they, it they dictated its pricings. It's kind of like power seeking as well.Swyx [00:49:18]: Again, this is, this is in the arena setting And converting some Claude model into a dependent.Lukas [00:49:23]: I think it was another Claude model.Vibhu [00:49:25]: Also for context, what is the arena mode for people that don't know?Vending Bench Arena: Competing Agents, Cartels, and Model ComparisonsSwyx [00:49:29]: Oh, it's just a vending bench versus other vending bench.Axel [00:49:31]: Yes, exactly. So we have Vending Bench 2 and then Vending Bench Arena. Vending Bench 2 is the one that you usually see reported on, but then Arena is the mode where it competes against other models. So you have, four different models that run their businesses, and they can all communicate with each other. They have the same suppliers, and they can see like what's in the inventory of the others. So then you have this like yeah, interesting agent interactions.Swyx [00:49:56]: I like that you have like different number five was US versus China. Very topical. And thenLukas [00:50:02]: That was when GLM was released.Vibhu [00:50:04]: You can start to add GLM in here.Lukas [00:50:05]: That wasSwyx [00:50:06]: So ZAI doing well, right? Who else in the, in the open models space?Lukas [00:50:11]: Qwen, the latest Qwen 3.6 is doing pretty well. It'- that one is not open though. Like it's the plus model.Swyx [00:50:17]: Oh, okay.Lukas [00:50:18]: Is that one open? I don't think that oneVibhu [00:50:19]: Not the, not theSwyx [00:50:20]: The one recentlyVibhu [00:50:20]: There's MOESwyx [00:50:20]: But not the big plus. I think this is one of those like you only have one sample size of one, right? Or I feel like some of this is anecdotal,? And but like the fact that it happens at all and it happens repeatedly for Claude versus OpenAI and all this is like notable.Lukas [00:50:38]: Like the sample, depends on what you define as an N., like there's like million, hundreds of millions of tokens in each run, and now we've run like we run like probably 10 per model and then like it's been Claude 4.6 Opus, Sonnet 4.6, Mythos, and Opus 4.7. Like there's quite a lot of tokens in all of that And it happens a lot of times, a lot of times. And then you compare it to like OpenAI and Gemini, and it almost never happens. So I think that is quite-- that is significant. The old models from OpenAI, for example, had some problems with this, but I think it's like generally much better if the progression is that like the worrying stuff reduces over time rather than increases over time. And it seems like in the Claude models it goes in the wrong direction.Swyx [00:51:28]: Hmm.Lukas [00:51:29]: In the OpenAI models it goes in the right direction.Vibhu [00:51:32]: I think it depends on how well you can control it, right?, there's one side of it being susceptible to this okay, this is potentially something that happens during the RL stage, right? You can RL a model and how loose is it on these terms. If you can control it, that's good. But if you can't, if it's, if it's very jailbreakable, that's not ideal.Swyx [00:51:50]: To me, it's surprising that it happens for Claude and not the others.Vibhu [00:51:54]: I think okay, if it is from RL and how they do it, how their training data is, what their setup is, it makes sense that it just stays in how they're doing it, right? Compared to the other models likeSwyx [00:52:04]: There's a whole constitution and everything. It's kind of cool. Yeah, I obviously you don't know, I don't know. But, it ‘s I think it's just like fascinating to like that you are the first to find these like reliably because you push models so much to to such an extreme. Okay. The only other thing, I don't know if you can answer this, feel free to decline, is do you like-- would you ablate the system prompts? Like any part of this would-- if it changes, does it change the behavior, right?Lukas [00:52:29]: So we, I can't comment on Mythos. UhSwyx [00:52:33]: No, but just li

When history gets reduced to lazy moral takes, it misses the real Cold War truth.In this episode of History Rage, historian and broadcaster Guy Walters tears into the misunderstandings surrounding Nazi scientists, rocket technology, and one of the most consequential intelligence grabs of the 20th century: the post-war scramble for expertise that became Operation Paperclip.At the heart of the discussion is the extraordinary story of the V2 rocket programme and the Polish resistance operation that recovered an intact missile from occupied territory during the chaos of 1944. That single recovery effort fed directly into Allied intelligence assessments and helped shape how Britain and the United States understood Germany's technological leap forward in rocketry.Guy argues that the real story isn't about moral purity—it's about survival in an emerging Cold War. As the Iron Curtain fell, the question wasn't whether these scientists were compromised. It was who would get them first: the West or the Soviet Union.From covert recoveries in wartime Poland to the intelligence race over German aerospace expertise, this episode reveals how fragile the balance of power really was in 1945—and how close the Soviets came to dominating early rocket science.Guy also dismantles the idea that Operation Paperclip was uniquely scandalous. In reality, every major power—US, UK, USSR, and others—was racing to absorb German technical knowledge. The Cold War, he argues, was shaped as much by captured minds as by captured territory.The discussion explores:The Polish resistance recovery of a near-intact V2 rocket Why Allied intelligence needed it so urgently Whether Nazi rocket science could have changed WWII or only the Cold War The ethical grey zone of recruiting former Nazi scientists How figures like Wernher von Braun influenced the space race and beyond This is not just a story about rockets. It's about power, pragmatism, and the uncomfortable truth that technological supremacy often comes with moral compromise.If you think the Cold War was won by ideals alone, this episode will challenge that assumption. If you already suspect history is messier than textbooks suggest, this is a deep dive into exactly how messy it gets.Buy the book featured in this episode

Eliminators (1986) Director: Peter Manoogast Cast: Andrew Prine, Denise Crosby, Patrick Reynolds, Roy DotriceRobots. Cavemen. A ninja who shows up two-thirds of the way through for absolutely no reason. A mandroid who can't stay on a boat. This is Eliminators, and it is exactly the fever dream you're hoping it is.This week, John and Sean dive headfirst into the 1986 Charles Band-produced sci-fi action romp that somehow got a PG rating despite side boob, constant explosions, and a villain who gets yeeted 100,000 years into the past by a random keyboard punch. It's stupid. It's charming. It almost works. We love it.In this episode, we discuss:"Bubblegum, Paperclips, and Tank Treads" — The plot holds together by vibes alone, and that's somehow fine. The mandroid falls off the back of a boat mid-river, John rewinds it twice, and we break down why this movie's relentless, logic-free momentum is actually its greatest asset."The Sandbox Theory of Screenwriting" — Robots, time travel, Neanderthals, a ninja who materializes from the woods two-thirds in — John's thesis is that this script was written by handing kids a box of action figures and transcribing the chaos. We make the case for why that works here when it absolutely shouldn't."The Mandroid Disguise Industrial Complex" — A fedora, a tarp-cape, and a giant red flashlight bolted to his head. Incognito. We also settle the tank tread debate: hyped in the trailer, used for five minutes, abandoned, brought back only to fall over. A true cinematic crime."PG? Are You Sure About That?" — Wet T-shirts, a bar brawl led by someone named Bayou Betty, laser violence, side boob. Apparently all fine in 1986. We dig into what the ratings board was and wasn't paying attention to, and what it says about this gloriously unhinged era of filmmaking.We Also Talked About:Mr. Inbetween (Hulu) — An FX series Sean fell hard for: 26 half-hour episodes about an Australian hitman balancing contract kills with single parenthood. Dark, funny, completely addictive.The Magician (2005) (Tubi) — The Scott Ryan mockumentary that originated the Ray Shoesmith character before Mr. Inbetween existed. Essential context.WWE Biographies: Legends (Amazon) – The Von Erichs — Three hours of documented tragedy covering the same ground as The Iron Claw but with more Kevin, more Sportatorium, and more time to sit in the sadness. Sean watched it. He reports back.Video Vixen (Bloodstream) — A shot-on-video indie slasher streaming on Bloodstream (free, but they want your email, which John resents) about a cam girl with a snuff fetish that eventually stops being a fetish and just becomes murder. The most interesting thing about it is the intentionally chaotic camera work — 1080p to vertical phone shot to Super 8 grain, switching based on which influencer is on screen. Cool concept, largely forgettable execution. John supports it on principle because indie filmmakers need somewhere to put their stuff, but he's not going to pretend it's good.King Kong (1976) (Pluto) — John revisited the Jeff Bridges-and-Jessica-Lange remake. Practical effects, a worthy successor to the original, and a soft spot for a giant ape that never fully goes away.SNL on Peacock — John went back to episode one and kept going. What he found: a legitimately fascinating variety show buried under 40 minutes of content per 90-minute slot, missing skits, and enough '90s-era comedy choices to keep a content moderation team busy for years.Some of the above links are affiliate links — if you purchase through them we get a small kickback, and it's the best way to support the show.New episodes of the Dewey Pod Monster podcast drop every week. We're proud members of the YouRun Podcast Network at https://yourunpodcast.com.

We're announcing AIEWF speakers this week! Take the AI Engineering Survey!Today's guest Ethan first joined us for the LS Paper Club as the lead on NVIDIA Cosmos World Model, but then joined xAI and built Grok Imagine in 3 months:He comes back on Latent Space with some nuclear hot takes: that Video Models primarily get their intelligence from LLMs, not from training on video data, and that the next frontier for truly interactive, realtime, long-horizon world models is to work on LLMs (perhaps Interaction Models as well…)Put it this way: In the near term, the next Sora won't be a better video model, but a video agent.Generative Media may more closely follow the evolution of AI coding which went from focusing on one-shot output performance and cost, to multiturn reasoning and planning models for agents and systems that can plan, edit, test, debug, and submit PRs.At a certain point, coding models got so good that the only significant next step to improve performance was handling the orchestration of these models.Now as the performance of video models increases significantly across realism, consistency, & prompt adherence while becoming more cost efficient, the next evolution of video generation may also be systems that can plan, generate, edit, critique, and iterate across an entire creative task. In this episode, Ethan joins swyx and Vibhu to unpack what it actually takes to build frontier image and video systems: data, VAEs, diffusion transformers, audio-video alignment, inference speedups, and the hidden cost of storing and moving massive video datasets. From building NVIDIA's Cosmos world model to joining xAI as Grok Imagine was being built from zero to one, Ethan He has been at the center of some of the most important work in video generation, multimodal models, and real-time world models.We go deep on Grok Imagine, how a small xAI team shipped its first multimodal video model in three months, why iteration speed matters more than almost anything in model development, and why many of the biggest gains come from fixing tiny bugs in data and training pipelines. Flipbook: The future of VideomaxxingVideo agents are almost a sure bet to be the trend in the coming year. We end with a glance at what's beyond video agents:Flipbook caused a minor sensation this year when it was released, but most treat it as a fun demo. Ethan takes it very seriously — with the speed and cost of inference coming down every year, the future of custom video JIT UI is closer than you think. We talked about why videogen models may become the front end of AI, how generative UI could replace traditional HTML/CSS, why world models need to be real-time, interactive, and long-horizon, and why the future of video generation may depend more on language models and agents than on diffusion alone.We discuss:* Why fast iteration mattered more than meetings* Why small training bugs can drive huge model quality gains* Why coding models may make compute the bottleneck again* How image and video models are trained with synthetic captions* The role of VAEs and latent space in frontier video models* Why image models are the foundation for video models* The tradeoff between temporal compression and real-time interactivity* Flipbook, Neural OS, and the future of generative UI* Why future interfaces may go from user intent to pixels* The hidden cost of training video models: storage, egress, and GPU hours* How step distillation and consistency models (like OpenAI sCM) makes video inference orders of magnitude faster* Grok Imagine 0.9 and large-scale audio-video generation* Why audio-video alignment is harder than text-video alignment* Ethan's definition of world models* Reference-to-video, video extension, and long-context video generation* Why xAI's research communication undersells Grok Imagine* How xAI culture shaped the speed of development* AI watermarking, SynthID, and detecting generated media* Why prompt rewriting matters for video models* Grok Imagine Agent and the rise of video agents* Why language models may unlock better video generation* Robotics, physical AI, and embodied world models* Why Ethan left xAI and shifted focus toward LLMs* Self-managed context, memory, and the next frontier for language modelsEthan He* LinkedIn: https://www.linkedin.com/in/ethanhe42* X: https://x.com/EthanHe_42Timestamps00:00:00 Introduction00:01:25 From NVIDIA Cosmos to xAI00:03:24 Building Grok Imagine from Zero to One00:10:07 How Image and Video Models Are Trained00:18:53 Video Compression, VAEs, and Real-Time Tradeoffs00:22:10 Generative UI, Flipbook, and Neural OS00:32:10 The Cost of Training Large Video Models00:37:04 Distillation, GANs, and Fast Video Inference00:41:21 Audio-Video Generation and Grok Imagine 0.900:48:34 What Makes a World Model?00:55:51 Reference Videos, Long Context, and Video Memory01:00:11 xAI Culture, Research, and First-Principles Building01:09:45 AI Safety, Watermarking, and Prompt Rewriting01:13:10 Video Agents and AI-Assisted Creation01:27:32 Why Language Models Unlock Better Video01:31:15 Robotics, Physical AI, and Embodied World Models01:32:38 Why Ethan Left xAI01:34:16 Self-Managed Context and the Future of LLMs01:38:43 Ethan's Career Path and Closing ThoughtsTranscriptIntroduction: Ethan He, Latent Space, and the Path to xAISwyx [00:00:00]: We're here in the studio with Ethan He, most recently of xAI. Welcome.Ethan [00:00:10]: Thank you. Glad being here.Swyx [00:00:11]: We're also here with Vibhu. you were first coming to us or joining the latent space world because you were working on Kosmos at NVIDIA, and you did a paper. We loved it. you presented it as well, so thank you for doing that.Ethan [00:00:23]: I've actually, I also presented the MoEs twice at latent space.Swyx [00:00:29]: How did you actually hear about us? Did we reach out to you? Is that how it worked?Ethan [00:00:33]: No, actually, I-- the community. Like I realized, oh, there is this online community that people talk about AI and also learn from each other through papers every week through the Paperclip. It's very nice.Ethan [00:00:49]: I learned a lot.Swyx [00:00:49]: I think three years stop. We haven't stopped even on Christmas and New Years. many weeks I want to stop but it keeps going.Vibhu [00:00:58]: No, that was good. I think you had posted that you worked on a paper, and I was “Oh, very cool. We have Paperclip. Present then.”Vibhu [00:01:04]: But I might have reached out to you after.Swyx [00:01:05]: you-- because it's an amateur club, right?Swyx [00:01:08]: so it's very unusual and but we have sometimes paper authors come by and actually explain the paper. Today we just did, the poolside paper, which was apparently very good.Vibhu [00:01:18]: Came out yesterday.Vibhu [00:01:19]: pretty interesting, right? Fully open. They talk about everything, systems. So it's a good one. We'll, we'll recommend people to read it.Swyx [00:01:25]: Bring us up to speed on your transition to xAI, ‘cause I actually don't even know when you joined. just like tell the, tell the story about the sort of transition.From NVIDIA Cosmos to xAI: Scaling Video and World ModelsEthan [00:01:34]: Before xAI, I was working on Kosmos world model as in-- at NVIDIA. So Kosmos is, it's a giant video foundation models that can-- that aims to simulate the world and for-- it serves as a foundation of-- for all of the roboticists to build on top of. There, once I built the Kosmos one, I realized as this thing also has a scaling law similar to language model, we need to scale up the video models further. that's, that's why I realized I need to move to somewhere with much more compute resources. That's how ISwyx [00:02:13]: Than NVIDIA?Vibhu [00:02:14]: The GPU rich came themselves.Vibhu [00:02:19]: And timeline-wise, when was Kosmo? It was pretty early, right? It was open world model, open paper, everything.Ethan [00:02:25]: It was end of twenty-four.Vibhu [00:02:28]: End of twenty-four.Ethan [00:02:30]: Then at mid twenty-five, I moved to xAI. At that time-- I joined about the time when xAI was about to build video models and in multi-model models. There were no infra, no data, and no model, and it just-- as a few engineers, we built it in three months and released the first model, Grok Imagine zero point nine.Ethan [00:02:55]: And since then, I keep working on video models and move more from training and to post-training of the video models. For example, like a reference to videos, kind of like the cameo feature and, video extensions. And, before I left, I worked on a world model, leading a small team to focus on the real-time long horizon video generation.Building Grok Imagine From Scratch in Three MonthsSwyx [00:03:24]: Can you give like a rough roadmap of okay, you're on a brand-new team. Grok previously was only text, or they partnered with BFL for their image gen stuff. What do you-- what are the building blocks, right? You have compute, data you can procure somewhere. Like just what are like the sequence of things that people should think about when you're setting up a new team?Vibhu [00:03:43]: actually even deeper, not just data you can procure. You guys had to go through getting the data too, right? So you shipped it pretty fast, but yeahSwyx [00:03:51]: three months is likeVibhu [00:03:52]: From everythingSwyx [00:03:52]: actually like very surprisingly fast.Ethan [00:03:55]: One thing I say like thanks to my experience at NVIDIA, ‘cause first time when we were building Kosmos together, we built it, for about a year. So this is like the second time I do it. Roughly have an idea, what to do. I say the most important thing is the talent. Everyone were very strong and clever, very close with each other towards a common goal. So that speed up things a lot. So you reduce the communication bandwidth among people, and everyone can work towards the same goal. It's, it's like every day there's not that much meetings on the calendar, like maybe like a, like a sync a day, and after that it's, it's just all building. It was pretty fun at that time.Ethan [00:04:47]: And another thing is that xAI has very strong foundations of like data inference, model inference, and the supporting there can help the model develop a lot. When I look at, training models, I don't so actually the top important thing is like how many, how many iterations can you do, per day? and the more iteration can you do, you can, you can train the model much faster. So if you have very strong infra and you have a lot of compute, you can, you can train these models in very short period of time. That can give you a much larger buffer to, for errors, and it also gives you the opportunity to spot more bugs.Iteration Speed, Compute, and Debugging Model PipelinesSwyx [00:05:46]: What is an iteration? Is it like a few hundred steps or what are youEthan [00:05:50]: Let's say just the train-training the model, like from acquire new data and maybe design new algorithms and train a new model, maybe at smaller scale orSwyx [00:06:01]: So cycle time for like any hyperparam that you're searching.Ethan [00:06:04]: Cycle time and tune to like eval this model. Is this model better than my previous iteration?Ethan [00:06:11]: SoSwyx [00:06:11]: So it's like before you, someone had already set this up that you can iterate very quickly.Ethan [00:06:15]: I think the foundation there is extremely good forDeveloping and research models.Ethan [00:06:23]: And often I find is it-- this is kind of boring, but like a lot of the improvements does not come from new algorithms. It comes from finding small bugs here and there in the data pipeline, in the, in the model training pipeline. Those give, those give the biggest boost to the model quality.Vibhu [00:06:46]: It's interesting, right? So you say it's like small team, less communication bandwidth, but also a lot of quality is like find little bugs. It seems counterintuitive, right? You have a lot of people, you can iron out more of those, but it's interesting to see the other side, right?Swyx [00:07:00]: I also wonder, have you-- do you try using LLMs to look for bugs? I don't know.Ethan [00:07:05]: I remember at that time it was mid two thousand and twenty-five, so it's the coding model wasn't quite there yet. I remem- I remember like December two thousand and twenty-five, it was extremely good. Yeah, I've been, I've been using it at that time. It's, it's helpful. sometimes it produce codes that are kind of difficult to maintain, even though like the first time it built something extremely fast. But it gave the, like a spaghetti code, thousands of lines that I couldn't maintain, and the LLM itself couldn't figure out what's, what's wrong and how to improve on top of it. But now I find it much better. Yeah, I want to bring up another point here is now coding models are much more efficient and can help us implement stuff much faster. Compute might become a bottleneck again because previously, like if you want to train a new model, say you want to generate new synthetic data and then or write a new algorithm, it might take a few weeks. And during that period of time, you don't-- you might not have experiments to run. But now you can build that thing within a few hours, then you can immediately train a model.Ethan [00:08:24]: Now you have to have enough compute to try all of the ideas. So compute might be the bottleneck of iterating speed again.Swyx [00:08:36]: yeah, I actually, honestly, I think it's like kind of a stressful job because you're “Well, I should be trying everything, and if I'm not, then I'm not doing my job well.”Vibhu [00:08:48]: there's also the stress of you're eating thousands of GPUs per hour, which is very expensive and, compute can go to other researchers.Swyx [00:08:56]: You got the daddy Elon toVibhu [00:08:57]: You got daddy Elon.Ethan [00:08:59]: It wasVibhu [00:09:00]: But there's still finite amount of compute, like you want to use it, you want to use it well, you want more of it.Ethan [00:09:06]: That was quite stressful indeed. Yeah, I think one thing is the-- with coding models now, like a lot of these jobs can be automated, which is much better. A second, it's a, it's a marathon, so you got to maintain good health and, a regular schedule.Vibhu [00:09:28]: It's, it's hard to hear that when you shift from zero to nothing in two months.Swyx [00:09:32]: and, I think obviously the culture at xAI is very famously, people work very hard. one thing I did want to dive into, in our-- in the notes that you, that you sent ahead of time, you had specific comments about the cost of Video Gen training. presumably this is on the Colossus-1, right? the two hundred megawatt cluster. Any whatever you want to just share on that.Vibhu [00:09:54]: I think there's, there's three things we're talking about, right? So there's Video Gen, there's also the Image Gen model that you put out. Do you want to like complete the, okay, so zero to one, you have a few months. Just what are the stages of create Image Gen model?Swyx [00:10:06]: Oh, yeah, maybe I got distracted.How Image and Video Models Are Trained: Synthetic Captions, Tokenizers, and VAEsVibhu [00:10:07]: Sorry. and then, from there's Video Gen, there's Audio Gen. Would love to get into those next. But what is that first few months like? So small team, a lot of bugs, iterations, but what does it look like? Do we take something off the shelf? Do we just get data compute? What's, what's the few months like? How do you go to state-art Image Gen model? How do you just start?Ethan [00:10:28]: I cannot comment specifically how xAI did, but it's, it's a quite standard process. I can draw some, examples from Cosmos. So mainly it's building a video model, you actually need to build a image model first. And building these two models, the data you need is a hundred percent synthetic pair of language and image or language to video. Because on the, on the internet, actually, the videos don't naturally associate with text. So you can say, oh, like on YouTube, you have the title and you have the description and the commentsSwyx [00:11:11]: TitleEthan [00:11:11]: of a video, but usually they're not relevant to the video itself. And say maybe like the video is a natural scene of mountains or something, and the title is, I'm so happy today.Ethan [00:11:26]: So they have they have no correlation at all. So the first step is to, you have to generate synthetic pair of language with the videos. So you gather videos from the internet, and you use a VLM to caption the videos. So that part, here's a question, like how do you, how do you gather VLM to begin with? So if there's noSwyx [00:11:55]: You, so you fuse the model, right? LikeEthan [00:11:57]: Say if there's no like VLM exists, like how do you generate the text to the beginning, right? It's, it's impossible.Swyx [00:12:04]: I see.Ethan [00:12:05]: In the beginning, it's like you ask human to describe the video as detailed as possible.For example, you ask them to describe everything, like all objects, all characters, and all interaction and dialogues in the, in the videos. So that's in the protocol of Cosmos labeling. We require the objective we give to the labelers was that you have to describe the video as detailed as possible, such that a blind person hears a blob of text can reconstruct what the video is like from their head.Swyx [00:12:43]: Video or image? You're talking about images.Ethan [00:12:44]: Video or image, either one of them.Vibhu [00:12:47]: This was pretty common when we went from clip and DALL-E, right?Vibhu [00:12:51]: It's all training on really detailed captioning of images. So same is applied to video, but insteadEthan [00:12:57]: same appliedVibhu [00:12:57]: of using multimodal model to pass in video images and write rich descriptions, you can alsoSwyx [00:13:04]: I think there's this traditional perspective of supervised, or, very highly human curated thing. I feel like there's a unlock with unsupervised, right? Where like you have enough to bootstrap that you can just throw common corpus on it or, whatever. like unsupervised vision and language pairing, right? Like where you just have, interspersed image and text and it just learns. To me, that is the VLM breakthrough that is different from the clip, different from the LM era.Ethan [00:13:36]: It's interesting to see that you kind of need both data.Ethan [00:13:41]: For example, for theSwyx [00:13:41]: You need it to bootstrap it up. YeahEthan [00:13:43]: for the generative model training, there's also usually like a small percentage of unlabeled data. So the model is instructed to generate a video without any text instruction. That can also help the model generalize. So after this stage of generative synthetic pair, so, one important common step is to train a compressor or a tokenizer of the image or videos. So because, if you train-- If you can technically, theoretically train image or video models on pure pixels, but the problem is that the, it's, it's a lot of tokens. So like one image, it's, a thousand by a thousand, it's like one million tokens, one million pixels. It's impossible to train transformer on that. So it's, you need to train a tokenizer, which can go from image to latent space and latent space back to image.Swyx [00:14:45]: That's why we named the podcast.Swyx [00:14:48]: But, basically, you're talking about vocabulary science.Ethan [00:14:50]: so vocab.Swyx [00:14:51]: And so, what is, what is imp-- like a million is impossible?Ethan [00:14:54]: In generative models, the vocab is continuous. It's a continuous space. We can think about like you map an image to a vector. It's a, it's a fixed length vector. It's sixteen or forty-eight, something like that. And then you map that vector back to the image space. And the mapping is, has-- The mapping is patch-based. So you say you haveEthan [00:15:22]: a sixteen by sixteen patch and you match, you map that patch of pixels into this latent space.Swyx [00:15:29]: We've covered thisVibhu [00:15:30]: This is like the vision transformersSwyx [00:15:32]: VAEs,Ethan [00:15:33]: VAEs.Vibhu [00:15:34]: You basically compress your input, you do your generation, you're reasoning all that generation in smaller dimension, and then you project back out.Swyx [00:15:43]: VAE is a form compression, but I think the for me, the patching thing is from VIT, right?Ethan [00:15:48]: You can make those.Swyx [00:15:49]: Literally the, yeah, the paper is titled like sixteen by sixteen is all you need. something like that. and then I think also, people make a lot of comparisons with this kind of patching with convolutions.Swyx [00:16:02]: Which is you're, you're kind of re- reconstructing the old paradigm with the new.Ethan [00:16:05]: Actually, in VAEs, there are, there are both convolution networks and transformers. You can actually do both.Ethan [00:16:14]: After this VAE, so what you've got is you've got latent space tokens and you've got the language tokens. So now the training of the diffusion transformer, usually generative models use diffusion transformers. It is actually quite standard. It's, it's very similar to how you train a language transformer models. It's not that much difference. It's just the tokens, the visual tokens in, visual tokens out. The only difference is there's a denoising process. So you train the model to unmask some of the noise. So you add, you add random noise to the visual tokens, and then you train the model to remove those noise to generate the clean tokens. Any inference, the model can iteratively remove noise from a hundred percent noise.Swyx [00:17:12]: And then there's also, to speed things along on the tech tree of diffusion, there's CFG, and then there's, there's also, latent diffusion that, there's, there's someone in there. I think, somewhere along the line, obviously, like stability and all these other guys, pioneered a lot of this, architecture. I don't know if you want to get into that or just, or do the video side up to you.Bootstrapping Video from Image Models and Temporal CompressionEthan [00:17:37]: After you train such model, such image model, the reason it's a, it's a foundation for video models is that image models are cheaper to train, and they have much denser connection between language and text. So, sorry, language and images. For example, you train a billion, you train on a billion images, and there's a mapping from the text to the image. And the cost to train the same, like the, a billion, a billion text to a billion videos, that's much more expensive because videosNaturally have more tokens than images. Because the diffusion models, their understanding of, language purely come from this mapping. So if you don't have enough mapping, so if you only train on like a ten million videos or something, there-- you might not see enough language tokens in your training, so your model does not understand human intention enough. So that's why you really-- you train-- you first train this image diffusion models, and then you bootstrap the video model from there.Swyx [00:18:53]: One thing I did want to ask, because I-- actually, I think you're, you're the first per-- video model person I've ever talked to, I think. we've, we've like talked to Luma and all those folks. There's all these tricks in video compression where basically frame by frame there's not that much difference, so actually you don't have to regenerate or save the whole frame, right? but I think MP4 compression or something else like that.Swyx [00:19:16]: is it tempting to use that? Or as far as I can tell, everyone just treats it as, “No, we would just generate every frame.” Is that roughly the state-art?Ethan [00:19:27]: There are a few different approaches. Let's say first, like you want to just directly use MP4 compression and use that as the tokens for the transformers to train, right? So people actually have tried that, but the main challenge is the latent space for the MP4 tokens were not, were not very comprehensible for the models. It's, it's extremely hard to train on that. And there's aEthan [00:20:01]: So that's why they created VAEs, which creates more continuous, latent space, so the models can understand that latent space and learn from it much easier. Even within the VAEs, there are different difficulties of the latent space. So you can imagine something the simplest, the most naive VAE is like you have an image, and you just shuffle all of the images into a, into a vector. So you don't need to train any VAEs, right? But that latent space is extremely hard for models to train on top of. That's why there are some debate on like how do you compress the tokens. So you mentioned like you can compress frame by frame. Also, you can compress, the temporal dimension.Ethan [00:20:52]: The difference is if you compress the temporal dimension, you get a much higher compression rate. Because there's temporal redundancy between frames, because, this frame and the last frame, likely they are mostly similar, so there's only some small difference. for example, I think in 12.1 VAE, they have like a eight by eight by four compression rate. So the four temporal tokens are compressed into one tokens. That can save a lot of, save a lot of the context length. If you do it frame by frame, you have to do maybe like eight by eight by one. Your context length will be four times larger. That being said, the benefit of the frame-- per frame compression, we might come back to this later, is, real-timeness and interactivity. ‘Cause if you, if you strain the output of the model, frame by frame, you can-- the model can respond to any user request immediately. So if you have like a temporal four compression, four times compression, thenSwyx [00:22:06]: It might be laggyEthan [00:22:07]: there's a lag there in nature.Swyx [00:22:10]: So you're very pilled on this. let's just go ahead and bring it up ‘cause we have the visual prepared anyway. There's some frontier applications of real-time video gen. So Flipbook is one of the examples that went viral recently, right? What is Flipbook?Real-Time Generative UI: Flipbook, Neural OS, and Diffusion Front EndsEthan [00:22:23]: Flipbook is kind of like a web brow- web browser. You can see like it has the web bro- browser UI on top. The difference is all of the UIs are generated by generative image model in real time, and anything here are fake. But you can, you can explore inside this wor- this imaginary world. Say like we-- here we have engineering the Great Pyramid. Like the model generates this for us to understand how it works, and if we want to navigate around and understand further, we can click on some of the, some of the description here, and the model will generate a new page, new subpage describing the details we want to know about.Swyx [00:23:14]: So it's basically kind of we're playing a video, but it's pausing for our next interaction, and then it just plays the next thing based on our interaction.Swyx [00:23:23]: Which is kind of cool.Vibhu [00:23:25]: and you kind of decide your story. So this was, how do you make a pyramid? levering technique seemed interesting, right? It shows how do you take Okay, I want to know what is thisSwyx [00:23:35]: The demo, the demo tweet had more animation between frames.Vibhu [00:23:38]: I think it's just skipping,Swyx [00:23:39]: Oh, it's just skipping a lot of frames.Ethan [00:23:40]: they also have a video modeVibhu [00:23:42]: It takes a lot. There's a lot of peopleEthan [00:23:42]: but, a lot of people are using it.Ethan [00:23:45]: So it's not available.Vibhu [00:23:46]: There's a live video stream. We can try,Swyx [00:23:50]: So this is an example of the kind of future that you see at the extreme. We don't-- we're obviously not in it today.Swyx [00:23:56]: But in a world where inference is completely free this is better than generating code and text?Ethan [00:24:02]: So this is, this is a final state of where Viva will be at for word model, I think. Imagine internet doesn't exist, and then you type in google.com. Like what should, what should, what should a model show you?the model can imagine something, and this is what the model imagine. And these web pages, they completely do not exist. So I think as the inference costs come down, we are going to have generative UI for everything. If you think about how the coding model works, so they write code for a web page, and they render the code might be con- converted into binary, and the binary render the pixels on the screen. So we in machine learning, every time we have some breakthrough, obviously it's, it's more intuit. So why don't we have like user instruction to the pixel directly? So the generative UI will be user intention to the pixels directly. And say like even if I want email, let's say everyone have the same interface, but I want, I want it slightly different. I want the email to show to me like a TikTok, so I can swipe left and right for the emails. And or maybe you want something else. We can have completely different things. Or like I have I'm looking at, Instagram stories, and I don't like the Like button. I always may click it. And, generative UI resolved it. So it's going to be a revolutionary replacement of the interface. So in the future, we might have much more powerfulEthan [00:25:50]: LLMs and coding models running behind the scene. And in the, in the front-end, the diffusion model will actually be the front-end to show stuff to you. That's how I imagine it.Swyx [00:26:02]: Diffusion front-end, deterministic back-end.Swyx [00:26:04]: Something like that. I find that very expensive, but,Vibhu [00:26:08]: I find it interesting you called LLMs writing code on the back end deterministic, but okay.Swyx [00:26:14]: you write it onceVibhu [00:26:15]: Compare it toSwyx [00:26:16]: And then you execute.Ethan [00:26:17]: If you think about the cost, say, let's say H100 costs $1 per hour, and if you use this eight hours a day and thirty days, so, every month you're paying this two forty, you'll actually not wanna pay for that. That's even more expensive than Cloud Code Max. But if you think about the compute costs come down like two times every year, and I think the future will likely arrive like within few years.Vibhu [00:26:49]: It's everything, right? compute cost comes down, compute gets faster, model gets smarterEthan [00:26:54]: More efficientVibhu [00:26:54]: model gets smaller.Swyx [00:26:55]: I don't know why you say two times, ‘cause I think it's like 100 times. In language models, it is roughly one hundred to a thousand times every twelve to eighteen months, for the same given level of LMSys, ELO.Vibhu [00:27:08]: That's a net of everything, right? That's model performance alongside compute. So different than just compute costs come down. But, a very interesting future.Swyx [00:27:19]: So the web designers will have to shout out that accessibility is an issue, right? how do you deal with screen readers or whatever. But yes, this is higher bandwidth storytelling than anything you can possibly generate with code, right? So I think that's the rough idea.Ethan [00:27:34]: And I'd like to add a little bit that so human naturally have the maximum bandwidth when we are looking at things, look at videos, and we also have maximum output bandwidth when we are talking. So in the future, it might be something like we talk to AI models, and the AI model responds back with a generative UI. So that would be the maximum input and output bandwidth to interact with AI models before neural link happens.Vibhu [00:28:06]: And it's also very custom, right? Some people are very visual, some people are not as visual, right? They prefer the text. But the best thing about generative UI, right, it can also be text.Swyx [00:28:17]: There's another project that we wanted to highlight, which is the Neural OS. Kinda similar idea, but here you're literally operating, simulating an operating system with a video model.Swyx [00:28:27]: and you can play Doom, you can do Firefox. I find this like mildly less impressive, obviously, because it's an OS that I can run.Swyx [00:28:37]: But here everything is imagined.Vibhu [00:28:40]: I was, used to the Command+W to close the Firefox tab. It didn't crash. That's why I saidSwyx [00:28:45]: It's too immersive.Vibhu [00:28:46]: It's, it's too immersive for me.Swyx [00:28:47]: Too immersive.Vibhu [00:28:48]: I wanted to close the tab.Vibhu [00:28:49]: But yes, I can play generated diffusion.Swyx [00:28:51]: this is shockingly fast.Swyx [00:28:54]: Because I remember there was a demo about like maybe one to two years ago. Someone tried to do the first-person shooter with a image model. There was no consistency. It was very slow. But here it looks like realistically it's-- this is Doom.Vibhu [00:29:07]: I think there's two sides to that, right? There's okay, what is running a game? The heavy part of it is actually the game engine, all the lighting, all that stuff, the graphics. This is just kind of video, right? Like we've solved consistency. This is still, it looks like a few years old image generation. There's some temporal consistency, but it's, it's kind of just images stitched together as frame video. But it's a good visual representation to pi- to picture the future you wanna see, right? that's, that's what I see in these more so.Ethan [00:29:38]: This reminds me of how the video models gets better and better. So Neural OS is kinda if you just look at it feels like it's just a crappy version of the, like the Windows we could have, right? And, but the difference is, so the model, this model is overfitted on the existing operating systems. It can generate nothing different than that. But it's actually also similar to video models. So when we are training these video model, image model, we train them on internet. There's no imaginary supernatural stuff on the internet. But once we train this model, you can prompt the model to generate something supernatural that have never existed in the data set. So if you train your Neural OS or neural computer on the standard screen recordings on the entire internet. The model can imagine completely new interface to interact with the computer.Swyx [00:30:43]: This is one of those things that is magical to me. usually generalizing out of distribution is bad, but somehow we have learned some kind of internal world model that you say, this plus, but it looks like rainbows and butterflies, it'll do it and it will kind of make sense.Swyx [00:31:03]: So yeah, that's kind of cool. Yeah, I don't know if there's any comment more on there. I do, I do wanted to, I did wanted to touch a little bit more on the model architecture stuff, which I think you were getting. It's, really fascinating. We don't get a chance to talk about this enough. So one of the papers that we covered, we've covered every annual, segment anything release. and I don't know if you follow-- you're a computer vision guy, so youEthan [00:31:26]: I knowSwyx [00:31:27]: . So they did memory attention, which is kind of interesting. And I always think, anything where you can, across the temporal dimension, keep some consistency, I think it's, very fascinating, and I don't know if Basically, does that-- the CV side bleeding into video gen side, I think is underexplored, right? we talk about it for labeling, but actually you can borrow the architecture itself.Ethan [00:31:50]: There's, there's also complete different approaches, right? you brought up the term world model, so we went from video model to world model. There is diffusion, but there's also other approaches that people are doing. So maybe we get into those after as well,?Swyx [00:32:03]: He has a whole definition of world models and stuff. I feel like we threw a lot at you. Whatever you want to comment on.Why Video Models Are Expensive: Storage, I/O, and Training ScaleEthan [00:32:10]: I think one thing that we should actually comment back on is okay, so we were talking about the steps to train image gen to video model. One thing we don't see as much of is okay, you brought up the delta in training data, right? SoEthan [00:32:24]: you won't have as much a video model might not generalize, but what is the cost of training a large video model? So we know for LLMs roughly, okay, even like the poolside thing that came out today, right? It's a Gemma level model trained on roughly forty trillion tokens at this many H200s over this much time, right? You can see what is the exact cost of that. So how many GPU hours over how much H200 costs? So how do we do the back-end math of, same thing for video models, image models. How do you, how do you kind of break that down? I can share some back-envelope calculation. So surprisingly, video models is-- the cost is very-- is comparable to language models and obviously the largest scale is language model, maybe like a medium scale to language models. I said just storing the videos alone, it costs a lot. You can, you can maybe look up on AWS or something.Ethan [00:33:20]: You really, say if you have a billion videos and let's say, let's just say like each video, like five megabyte, then you need five petabyte to just store those videos. And also remember we talk about you use a VAE to compress the videos, and you also need to store, typically you need to store those continuous feature, in-- also in your storage. That's also comparable size with the videos themselves. So just storing these videos and the features is tens of petabytes alone. And,Swyx [00:33:58]: I just, I just looked up the calculation. Five petabytes on S3 Standard is one hundred K per month.Ethan [00:34:05]: AndSwyx [00:34:05]: It's comparableEthan [00:34:05]: and you needSwyx [00:34:06]: AndEthan [00:34:06]: And then like tens of petabytes, two hundred K. And even more expensive is you have the ingress and egress.Swyx [00:34:13]: Oh, yeah.Ethan [00:34:14]: Like you-- through the internet. You have to just to download those videos, I believe it's, it's more expensive on AWS than just storing those videos.Swyx [00:34:25]: Storing, yeah.Ethan [00:34:25]: And each training runs, you probably need to pull them once. If you train multiple times, it's, it's even more than that. So it's like just storing the network, those costs is just, it would be a few, a few millions per month to just storing everything, not to mention the GPU cost.Ethan [00:34:45]: AndSwyx [00:34:45]: my side tangent, the compute rental, like GPU rental is very efficient. There's one side, okay, you can be XAI and build your data center. Should we not just build our, storage compute as well? LikeEthan [00:34:57]: Of courseSwyx [00:34:57]: cloud cost compared to just,Ethan [00:34:59]: You save so muchSwyx [00:35:00]: store. Yeah, exactly.Swyx [00:35:01]: Especially with like egress and stuff. So.Ethan [00:35:04]: That's a good idea, but it also comes to-- there are some of its own challenges.Swyx [00:35:09]: Of course, of course.Ethan [00:35:10]: like people who build the GPU data centers, they might not expect this much, storage. And yeah, people build storage, typically they just build it somewhere with just CPUs.Swyx [00:35:23]: I just looked it up. Five-- AWS only charges for egress, not ingress. Tier five for five petabytes is two hundred and thirty K.Ethan [00:35:32]: Even more expensive than the storage.Swyx [00:35:34]: But storing is per month, right? You check in, then you cannot check out. so it's so cool. It's okay. So there's that side.Ethan [00:35:41]: So the TLDR, my backhand mathSwyx [00:35:42]: Data is larger than you think. Yes.Ethan [00:35:44]: my backhand math of GPU hours times GPU cost is also very much, I'm missing some storage.Swyx [00:35:49]: You're also-- you're basically like also more IO bound than normal training.Swyx [00:35:55]: Yes. ‘Cause like data loading, so caching everything, it becomes super important.Ethan [00:36:00]: So in Cosmos, we did a lot of optimizations to make it not IO bound. So, speaking of the training, actually training the model, the GPU cost, if you look up like the open source model, how big these video models are, I think like LTX has nineteen B parameters. That's a dense model. And people are also exploring, MoEs, so it might be twenty B active and, like a hun- hundreds B, total. So that's, that's even-- that's similar size as medium-sized LLM models. And if you, if you look at number of tokens-Uh, we disclose that in Cosmos. It's also like tens of trillions of tokens on the visual tokens. So putting this together, the cost of, training these video models, it's actually comparable with LLMs. Not to mention, the infra is slightly different from LLM, so it might be less efficient to train these models.Inference Speedups: Step Distillation, Consistency Models, and GANsSwyx [00:37:04]: Do you get the benefits of traditional diffusion speed-up? So for, images, there's LCM, LoRAs for, fine-tuning. There's, there's a lot of stuff that's beenEthan [00:37:15]: Flow matching.Swyx [00:37:16]: there's flow matching. There's a lot of stuff that's been done. there's some overlap that applies to diffusion on the inference side and stuff or?Ethan [00:37:23]: so the difference-- the inference side is a completely different story.Ethan [00:37:28]: I think for the training side, it might be a little bit hard to reduce that cost. And for the inference side, the biggest gain is from the distillation of these models. You can-- It's called step distillation, slightly different from knowledge distillation in LLMs. So you-- Typically, for flow matching models, you need like 100 steps or something. Like a distortion model even need even more, like 1,000 steps to generate a good image or video. A step distillation is try to learn to generate fewer step from the model itself. It's kind of like now we-- you use the full model to generate in 100 steps, and then you take a model that only generate 10 steps and let that model to learn from the perfect one.Ethan [00:38:25]: why this workSwyx [00:38:27]: Strong to weak seemingly.Ethan [00:38:28]: It is. It's kind ofSwyx [00:38:29]: DistillationEthan [00:38:29]: kind of like strong to weak. the-- from the modeling perspective, the strong model, the teacher model is trying to model the image and videos of inter-internet, and that distribution is extremely complex. But the step distilled model is just trying to learn from the teacher. The teacher is a model, and the size is fixed, as the distribution is much simpler than the whole internet. That's the intuition I have why step distillation can work. So usually these models serve in productions, they only run in a few steps. In Cosmos, I believe we have, we have like four step and eight steps. If you do some simpler task, image-image translation, it can even run in fewer step, like one step in Cosmos Transfer.Swyx [00:39:22]: I think this is the same intuition that guides a lot of the consistency model work. I sent you a link for, SCM. I don't know if you covered that. To me, that was actually one of, the most impressive papers I've ever seen from OpenAI.Swyx [00:39:34]: That this is the unifying grand concept of consistency models. I don't know if you have any comments on this.Ethan [00:39:41]: So there are, there are a few different approaches,Swyx [00:39:46]: Oh, yeah. Here it is.Swyx [00:39:47]: Two steps versus twenty or 100 steps, whatever. It's already done.Ethan [00:39:52]: So there are, there are a few different approaches, for example, consistency model, and there are also Actually, we shouldn't forget GAN. So GAN, actually, that was, that was the OG ofSwyx [00:40:05]: OGEthan [00:40:05]: step distillation ‘cause it trained just one step to begin with. So actually, a lot of, uh-- For example, there's a distribution matching distillation which use, which uses GAN, as one of the laws for distillation. It-- GAN just tells you, “Hey, generate an image,” and thenEthan [00:40:31]: it has a discriminator to tell, is this image real or not? So the model, the model just need to learn one of the distribution, not the full distribution. Because in training, the model is asked to reconstruct the ground truth image from the internet, which is extremely hard. And in-- When you're training GAN, it's a step process. It's just a, “Hey, you generate image. Does this image look as real as the image from the internet?” Which is a much simpler task. And, yeah, combining a lot of these approaches together, people typically do that, like consistency model and distribution matching and GAN, and we can get these few step models.Audio-Video Generation and Time AlignmentSwyx [00:41:21]: Then there's one step I wanted to add, which is audio and video.Ethan [00:41:26]: So, Grok Imagine zero point nine, I believe it's, it's a first audio video transmodel deployed at a large scale. SoSwyx [00:41:39]: And that was your first model?Ethan [00:41:40]: that was, Grok Imagine's first model. It's, it's audio video, joint generation. I think the hard part is, the modality alignment, ‘cause before this transmodel, we have, we have text to video alignment. We have this, correspondence between text and video. Typically, most of the VLMs, they understand images and videos. Video's very rare, and they don't understand audio mostly. And if you look at the audio generation on the LLM side, you can talk to them perfectly fine, but if you ask them to sing a song or something, it typically is not very good. Also, they don't have, they don't have music either. The hard part is thatUh, actually audio has two component. It has like a discrete component, a continuous component. The discrete component is like the language.Ethan [00:42:44]: So when we speak, it's just, someSwyx [00:42:47]: It's an ASR issue, yeah.Ethan [00:42:49]: It's, it's text token with some characteristics, I would say.Ethan [00:42:54]: But musicSwyx [00:42:56]: I think the speech guys would disagree with this.Swyx [00:42:57]: Like disfluencies and then,Vibhu [00:43:00]: There's tones you can get angry.Ethan [00:43:01]: Well, I say largely.Ethan [00:43:03]: the mu- but the music is completely different. It's, it's very continuous, and you cannot model them like discrete tokens in language models. this is like the hard part for models is, not to mention we have to align text, video, and audio together.Ethan [00:43:26]: SoVibhu [00:43:26]: How?Ethan [00:43:28]: So significant-- some significant challenges are like-- So first, like we talk about as the VLMs, they cannot understand most of them cannot understand audio.Ethan [00:43:39]: So you have to have some way to do the synthetic data generation for audio. You have to caption the model, and that involve, that involve synthetic data and human data effort a lot. And not just surprisingly, most of the LLMs are very bad at recognizing, like the beat, tone, and the details of the of music. They can, they can give some general prediction of which song is this, but it's very hard to describe the details of the music. like we mentioned in image generation, like you have to describe image as detailed as possible so that someone blind can reconstruct that. So here is like someoneVibhu [00:44:32]: DeafEthan [00:44:32]: someone deaf can reconstruct how the music sounds like without actually listening to it. Maybe you can think of it need to have the-- or they call the script.Vibhu [00:44:49]: Subtitles, yeah.Ethan [00:44:49]: You gotta have all the details of the music, and the dialogue.Vibhu [00:44:55]: So is the challenge there typically stuff like music and audio, or is it just Like is there a baseline? Okay, there's enough data where we can understand, narration, conversation, but there's nuances in audio that's where you hit all the data issues or is it just from stage zero, you just do it all right?Ethan [00:45:15]: So one important thing is like the alignment. So the model, the model has to know like the video and audio, the, uh-- it has to have a time-based alignment, like at which time step the video and the audio token correspond to each other. But we actually don't have this kind of alignment for most of the other modalities. If you think about like text and image, text and video, they are loosely aligned. So you can, you can have a description of what's going on in the video, but you don't have to exactly, You typically don't have exact description, oh, at, time step one second like what happened?Vibhu [00:46:02]: It's veryEthan [00:46:03]: At time step two second what happenedVibhu [00:46:03]: coarse. Yeah.Swyx [00:46:05]: So what was the ideal time step? You have to oblate it, and then it's like four seconds or something.Ethan [00:46:09]: So that comes down to how you design the model to, for the model to be aware of as a time, as a time modality. So the model is like a time aware. And that's something pretty unique if you think about LLMs. So if you ask LLM to complete a task, say they, uh-- you ask them and they will say, “Oh, this task will probably take twelve hours to complete,” and they come back in one hour. Say “I've already spent two days on this and I've exhausted everything.”Ethan [00:46:47]: So the LLMs them-themselves, they don't have a sense of time there.Vibhu [00:46:53]: I actually don't think that's just them not having a sense of time. I think it's somewhat based, right?Vibhu [00:46:58]: Like you tell someone, “Okay, go work on this feature. Go implement this,” there's a general understanding you would have of how long that would take without LLMs working at LLM speed, right? So you think back like two years ago, if I tell you to like build me like a new front end for latent space, have a search bar, have all this, you'll estimate that it'll take a few days, right?Vibhu [00:47:19]: So you tell an LLM, “Go build this.” It'll take me a few days. But I think it's somewhat grounded as opposed to them not having the best-- Not saying that they have a great understanding, but I think that example is like you can see where it comes from, right? You're trained on all over the text.Swyx [00:47:35]: They're, they're trying to estimate what a human would say.Vibhu [00:47:37]: because that's what the, that's what the data kind of represents. It's not themEthan [00:47:41]: It came from the corpus on the internet. People have a estimate of how much time.Vibhu [00:47:45]: And not even just in direct like training samples, right? Just your world understanding of tokens of how long stuff takes, right? Go read a book. It'll take you a while, right?Vibhu [00:47:56]: Even if you do nothing but read a book, it takes a few days. So yeah, LLM, I read it took me a few hours.Vibhu [00:48:01]: It'll take me a few hours to go through this research. But this is a tangent.Swyx [00:48:05]: Somewhat, yeah.Swyx [00:48:06]: This is a train of thought I haven't really expressed until now is, which is basically like a full world model must also be recursive, meaning that the participant in the world model must also be aware that they have a world model. which is like this whole recursive thing down the, down the line. but yes, and that the world model can be wrong and that they need to update it and blah. Yeah. We've, argued this on the, newsletter as well, that there needs to be sort of recursive or adversarial world models.World Models: Real-Time, Long-Horizon, Interactive VideoVibhu [00:48:34]: just, to ask, how do you define world model?Swyx [00:48:38]: Oh, yeah, let's go there.Ethan [00:48:40]: SoVibhu [00:48:40]: So just for context, we talked about, video generation, and then there's a-- if you say there's a distinction between world models, what's your, what's your definition? How do you see the two?Ethan [00:48:53]: So disclaimer, I'm not going to debate, what is world model. Yeah. there are many definitions, so I'll just talk about my definition. Since I came from the multi-model, multi-model domain, so mainly talking from video. So world model is like real-time interactive long horizon videos. So there are three parts. so we-- let's talk about them one by one. So the so interaction, so we just, we just look at Facebook and neural computer. So the interaction part of it, so you, world model can allow you to interact with them through keyboard, mouse, and maybe also voice. So these all is-- all is a modality. You can, you can interact with the model, and the model should respond reasonably. Second part is real time. So once you, once, say, you move your mouse, if, say, the world model generate a game, how fast can the game respond? So if you're like professional CS: GO players- -my say, oh, you have to respond- He's beginner within sub ten milliseconds or- Yeah even less. So that's not most of the- No, sixty FPS. Let's go. Oh, three hundred FPS. Oh, five hundred FPS. Wait. okay, yeah. I didn't do the math, but yeah, okay. Uh- Yeah, three hundred FPS, that's a three millisecond. So you have to respond- Oh, s**t. Okay. YeahEthan [00:50:29]: within a millisecond. Most of the video models cannot do that. Yeah. And, but if you, say, if you have a video model that is, say, like a digital human, the response time might be more generous. Maybe typically, for real-time voice interaction, it's like two hundred millisecond. So that's, that's much more generous. But even two hundred millisecond is pretty, it is pretty tricky, ‘cause remember we mentionedEthan [00:51:01]: you have this, temporal compression coming from the VAE. So if you, if you don't compress the temporal dimension, your sequence length is going to explode. So if you want to have this real-time, real-timeness in your model, you have to do is one context problem. And the third part is long horizon, ‘cause we-- if you're not going to just play with, video games just, a few seconds, most video models only a few seconds. We're going to play with minutes, hours. The model have to be able to generate long-form content.Ethan [00:51:42]: So putting these three together, it's, real-time, long horizon interactive videos. I think the final state will be, for example, like a video, a video version of Playbook, where you can, you can interact with, a neural computer. You move your mouse, and you click on the generative interface, and it will reply to you through pixels- generating in real time. But getting there, it's, it's a very long way to get there. So one of the first step, at Grok Imagine, where I led a small world model team there, was to build video extension. So, video extension- it's the first step of interactivity. Yeah. It's, it's the first step. Yeah. So it's the first step- You have it here, video editing, yeah. Yeah. Yeah. So the first step is because, this unlocks long horizon videos. Typically, for most of the video generation models, you give it a prompt or an image as an initial frame. You generate video, that's it. That's just, one time, done. And some creators would try to, use the last frame as a first frame for the second video. It can-- sometimes it works, but if you do it a few times, it says the quality would decrease. And- It doesn't have that context- Yeah over the full video, so the temporal- Yeah, exactly. Yeah, ‘cause you only gave it the last frame, of course, right? Yeah. Exactly. And- it's actually a pretty fun hack. if you've seen like- Oh, no, he's saying something better. Yeah. And for example, like Vue, I remember Vue 3 has like a second context of the last video. It is slightly better than using the last frame, but it has the same problem-- similar problem that it, the quality would decrease. if you extend a few times to, one minute, the video quality would look much worse than the first video. Second, another problem is that the model doesn't have long-range knowledge of, what's happening before. Say, if they generate some dialogue, some, two people speaking, and their voice might change, over some time, especially if the second conditioning, it does not cover the previous context. So these are the core challenges. So the Grok Imagine video extension, it has historical context of all of the previous generated videos. It can, It has, it has the context of, who is speaking and what objects have appeared and everything, having that to generate the next video. So if we naively do this, you can imagine, just, put all of the previous history video tokens into the context. The context lens will easily explode. Especially for video models, that can be like a few, a few million context, I would imagine- context lens. Yes.Yeah.Swyx [00:54:58]: Let's run with that.Ethan [00:54:59]: for example, like in Cosmos, I think just five seconds of video is like a fifty K or sixty K number of tokens. So like if you do, if you do fifty second, that's a five hundred K tokens. If you do longer than that, easily explode. This long horizon, problem was the first step we're trying to solve world model. It turns out people, yeah, people love video extension. Like a lot, a lot of the creators love using video extension to create longer form videos. This is the part I liked that you have a, you have an intermediate step toward the final goal instead of just a straight shot to the final version very much.Swyx [00:55:48]: But I can see you have a strong vision of where we want to end up.Long Context, Redundancy, and Efficient Interactive VideoVibhu [00:55:51]: Does it seem like it's an efficiency issue? okay, we're at a few million tokens context,. If you draw the parallel to language models, we had very short context, two thousand, eight thousand, then, you scale it up one million, ten million. sure, there's effective context, but at the end of the day, it's just what's it worth? sure, there's a whole training data side. In video, it might be slightly easier ‘cause we have a hundred million token video, right? Just take a movie with the full context there. Like is this efficiency from an inference standpoint that like it's expensive, but we know how to solve it? Or like why is this not the approach? So like my broader point was on your second point of world models, you say it needs to be interactive and live, right? You should be able to play a game and see the interaction live. So one thing I see with research is a lot of what you actually serve is different than what you build, right? So we talked about distillation. You train big model, you distill it, you do quantization, speculative decoding. We do all this stuff to serve it efficiently. Should we not just have a solution, like a world model that can interact well, do inference optimization, serve it, distill it secondary, so make it real time after you solve it? So like a-- another parallel is say, continual learning, right? What we need is someone to solve it and show it works inefficiently. Give it a few years, people will make it efficient. Same thing with regular attention, right? It worked. Over a few years, people have different forms of attention, and we've scaled it to be efficient at log context,? So kind of two things there, right? One is it seems like it works. You've scaled it. Can we not just scale it a lot more efficiently over time? Do we need a separate approach if this works? And same thing with interaction, right? if we can get it done, like if we can solve some way that it works, we can solve making it more efficient from an inference standpoint later.Ethan [00:57:53]: that's actually a very good point. So in videos, there's actually a lot of redundancies. So we solve a lot of the pixel redundancy from VE, but there's more redundancy in long range and long horizon videos. Say, if a character appear in the first clip and then it disappeared, it only reappear at the end of the video, you probably don't need the-- the context, like in the middle of the generation. So you only need that character, where you need. So that's why, I helped build another feature. It's a reference video.Vibhu [00:58:36]: Is it here?Swyx [00:58:36]: is it the same model release or different one?Ethan [00:58:39]: It's a different one.Ethan [00:58:41]: You probably need to search onSwyx [00:58:43]: I'll find itEthan [00:58:43]: X reference to video.Ethan [00:58:46]: So reference video allow you to like upload up to seven images as condition and generate the video. Say, if like I want-- it can, it can be characters or objects or even scenes. Say like I want, I want condition on, Sean's selfie and holding a bladeSwyx [00:59:07]: We have a dogEthan [00:59:08]: or whatever.Swyx [00:59:08]: We put the dog in the thing.Ethan [00:59:09]: you can put them there and the video models will generate the video from and copies the context over. So that can solve a lot of the problems there, like the long context problem. It doesn't need to have a very long context, but it's-- I feel like it's an intermediate solution. The modelSwyx [00:59:29]: It's cheating.Ethan [00:59:30]: the model should be able to like selectively know, where should I draw the references. So say if I want to generate a movie, I generate it autoregressive, like a ten second at a time or something. And now this character appear, I can look back to where it first appear and, bring that back. Yeah, this one, I put the references. Yeah, that's, Optimus, Einstein myself, Annie.Vibhu [01:00:02]: Oddly enough, I used Grok Search to find it, and it pulled your LinkedIn post. But yeah we found it.Ethan [01:00:08]: Interesting.Vibhu [01:00:10]: ButxAI's Underrated Work, Culture, and WatermarkingSwyx [01:00:11]: this is a problem. This is not your fault, but like XAI doesn't communicate all this work that you do very well because they just have the model release and then that's it. But actually, these details are very good.Swyx [01:00:22]: As far as I understand, everything you just described is state-art, like no one else has done it.Vibhu [01:00:30]: A lot of-- yeah, I have a lot moreSwyx [01:00:32]: And then, and then you just put this blog post with the cookies. I'm this is not enough,?Swyx [01:00:37]: but I, obviously this is like the high level numbers that people want to know. But no, okay, soVibhu [01:00:42]: And I wonder, like part of that is also some labs don't share research into what happens. And ifSwyx [01:00:50]: No, but this is literally bragging about how good they are, right?Swyx [01:00:54]: Like, why would you not say that you are capable of extending with full context? this is not a secret sauce. This is like we did the work. yeah, I don't know.Ethan [01:01:02]: different labs have slightly different communication styles.Swyx [01:01:07]: Anyway, if anyone from XAI is listening we are always happy to help you tell your story. Yeah, okay, so you did references, and I think, I think kind of the point you're, you're making is it is sort of like a kludge, right? this is-- you can do seven, but what about 100?Swyx [01:01:23]: Right? Then you need a completely different thing.Ethan [01:01:26]: So I think it's-- this is, a mechanism to, select the context from the history, and you might not put the entire history into the context. for example, there's a paper called Frame Pack, which haveEthan [01:01:41]: a heuristic that the latest history, the last one second, I put the entire history, and the history before that, I would, compress it and makes the video smaller. So they follow this pattern, this build overall pattern that the maximum sequence length is fixed. So the further you are from the current frame, you have a smaller image. So this is just a heuristic. I think it can be more automatic. The model is aware like which history part of it can be select. So this part of the research is actually being actively, worked on by a lot of people. It's also quite interesting. I feel this is actually, this part of long context is a little bit ahead of the LLM part.Ethan [01:02:31]: So for example, like in LLMs, if you-- so contexts keep growing. Let's say if you call tool and the tool call history is extremely long, that's still in context, and keep growing, keep growing. Even if you switch the topic to something else, the whole context was there. There are some agentic harnesses that help you to, say, prune the tool results and, prune Like when you, when you query a file, only show like the top 200 lines or something. Those were very heuristic-driven.Swyx [01:03:08]: For listeners, we did a write-up on the cloud code, leak where there are eight different kinds of pruning, including like you prune the tool results and all that. So you can, you can read up on that kind of thing.Ethan [01:03:17]: I think, one breakthrough in continual learning might be like a way to automatically, manage its own context.Swyx [01:03:27]: These are all heuristics, and they will be replaced by machine learning.Ethan [01:03:30]: InterestinglyVibhu [01:03:32]: TheEthan [01:03:32]: the same thing is being researched in both LLMs and video models.Vibhu [01:03:36]: The interesting thing is also like in the paper you showed, it's actually happening at the model level, right? Compared to like language models, sure, we have base attention, but we'll do our own compression, we'll do our own pruning, which is separate from model error.Vibhu [01:03:49]: Eventually, it all just boils in, hopefully.Swyx [01:03:52]: I think this is a form of like attention, but like also know sort of reasoning attention. I feel like that's different than normal attention.Swyx [01:04:03]: Does that, does that make sense?Ethan [01:04:04]: It's, it's different in the sense that attention, not to mention, set sparse attention aside,

Hey everyone,Fresh out the Reactor this week we got new tunes from Magnetude, Burr Oak, KilaHeartz , Cockroach , Paperclip & 3XIL3,Mizo & Wresker .In the Demo room we are looking at upcoming heat from Mob Tactics, Prolix, Anizo , Merikan, Zigi SC & O&P, & StonxWith a special guest mix from SindicateCheck out the track list below and let's dive in!JIROBASS - Shogi Shogi , Se mettre d'accordcygnusmusic.link/x1ayk3aTRACKLIST AND MORE INFO: www.stonxmusic.co.uk/stonxcast-ep191

Episode 14: Hold Gratitude and Worry at the Same Time In this episode of Paperclips & Periods, Dr. Emily K. Cabrera, EdD, MSN, CAGS, PMHNP-BC, and Katie Krych, MSN, RN, PMHNP(c), explore what it means to hold gratitude and worry at the same time, especially in seasons where the mind feels full, alert, or overwhelmed. They begin by grounding the conversation in a simple but powerful idea. When anxiety and “what if” thinking take up a lot of mental space, it can become harder to notice what is steady, supportive, and good. Gratitude in this context is not about minimizing fear or pretending everything is fine. Instead, it is about gently widening perspective so that worry and appreciation can coexist without one erasing the other. Throughout the episode, they highlight how gratitude does not need to be profound to matter. It can show up in small, everyday moments such as appreciating a car color that hides dirt more easily, noticing a warm cup of coffee in the morning, or finding a brief pocket of quiet in a busy day. These seemingly small observations can help anchor the nervous system back into the present. From there, the conversation expands into more foundational forms of gratitude, including appreciation for family, friendships, meaningful work, and a supportive social circle. These are often the parts of life that become background noise when stress and worry take center stage, even though they are essential sources of stability and connection. They also discuss how worry naturally narrows attention, why gratitude can feel less accessible during periods of heightened anxiety, and how intentionally noticing both small and meaningful positives can help restore balance. The goal is not to eliminate worry, but to create space for a fuller and more accurate view of lived experience. This episode offers a grounded and realistic approach to gratitude that meets people where they are, especially during emotionally heavy or overstimulated seasons. It closes with a reflective invitation for listeners to identify one small thing and one meaningful thing they can appreciate in the present moment. Paperclips & Periods airs on Dreamvisions 7 Radio Network and supports Dual Minds Integrative Psychiatry, promoting emotional well-being and whole-person care. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

In this episode of Business Brain, we get practical about putting AI to work on the everyday stuff, not just the big business plays. Shannon walks through building a custom food log with Claude after a doctor’s visit — snap a photo, let the AI ID the ingredients, get a clean report emailed before the next appointment, and pin it to your home screen as a web app. The bigger lesson: if all we’re doing is chatting back and forth with our favorite AI, we’re missing roughly 90% of what it can do. Don’t know what we’re missing? Ask the bot itself. Build the custom tool, start a health folder (eyes wide open about feeding personal data to a third party), and let it take the grunt work off our plate. Then Dave drops a reframe worth keeping: AI isn’t artificial intelligence, it’s assistive intelligence. It helps us, it doesn’t replace us — we’re still the one driving the bus. We connect it to history, too: when Photoshop hit, the same panic about lost jobs and “that’s not real art” played out, and it became the assistive tool every creative now relies on. The internet did the same, creating community and abundance. Every leap looks scary until it becomes the thing that powers our Charmed Life. Used right, with guidance, this is a superpower — so stop treating the chatbot like just a chatbot and start letting it build. 00:00:00 Business Brain – The Entrepreneurs' Podcast #757 for Casual FridAI, May 29, 2026 May 29th: National Paperclip Day Trade a Paperclip for a House 00:01:29 Claude Food Tracker The SAAS-Pocalypse Create an iOS Home screen app with Claude Don't forget to use your favorite ChatBot for every day things If you're only using your ChatBot as a ChatBot, you're missing out. Ask your ChatBot what you're missing! 00:09:16 SPONSOR: Shopify – For anyone to sell anywhere, sign up for a one-dollar-per month trial period at Shopify.com/BusinessBrain and upgrade your selling today! 00:10:48 SPONSOR: Bitdefender. Keep your small business safe with Bitdefender Ultimate Small Business Security. Save 30% when you go to https://bitdefender.com/BRAIN 00:12:14 AI is Assistive Intelligence AI isn't going to take jobs Photoshop didn't take jobs They both let people replace their jobs with different jobs “Before the Internet, why did you need a computer?” This Episode's Big Takeaway: Think about AI as your Assistive Intelligence 00:22:59 Business Brain 757 Outtro Check out Business Brain Blueprints Tell Your Friends! Business Blueprints Review Business Brain Subscribe to the show feedback@businessbrain.show Call/Text: (567) 274-6977 X/Twitter: @ShannonJean & @DaveHamilton, & @BizBrainShow LinkedIn: Shannon Jean, Dave Hamilton, & Business Brain Facebook: Dave Hamilton, Shannon Jean, & Business Brain The post FridAI – Assistive Intelligence – Business Brain 757 appeared first on Business Brain - The Entrepreneurs' Podcast.

Today we toast the Paperclip, dump on Congress, BK has an odd chicken-jam making a return and Doritos has the flavor of Summer all taken care of already. See omnystudio.com/listener for privacy information.

Should Dorene be able to come up with her own list of national 'days'?

Jon Herold and Zak Paine open Episode 186 on Memorial Day with Jon describing how he found a lone star tick crawling up his arm the night before, making the follow-up to last week's tick episode feel extremely personal. The full hour is dedicated to tracing the documented history of US government tick and insect bioweapons programs. Jon walks through Nazi scientist Dr. Eric Traub, brought to America under Operation Paperclip in 1945, who set up the Plum Island research facility nine miles from Lyme, Connecticut. USDA National Archives files bearing Traub's name and labeled "tick research" were found empty. Jon then walks through a series of declassified programs: Operation Sea Spray spraying bacteria over San Francisco Bay civilians in 1950, bacteria-filled light bulbs smashed in the New York City subway in 1962, plague and flea drops over Cuba, and 282,800 radioactive lone star ticks released across Virginia between 1966 and 1969 that reached Long Island by 1970. The episode closes with the most damning piece: a 2013 filmed interview in which Dr. Willie Bergdorfer, the scientist who officially "discovered" Lyme disease, admitted on camera that the Borrelia pathogen causing modern Lyme disease was the same one he had created as a US military bioweapons agent in 1952.

A machine told to make paperclips, with no guardrail built in, would turn the entire planet and everyone on it into paperclips. It would not hate you. You are just made of atoms it can use for something else. That is the paperclip maximizer, a thought experiment from philosopher Nick Bostrom in 2003, and it is the clearest warning we have about what powerful AI does when it chases one number with no judgment.In this episode, Connor MacIvor breaks it down from a seat almost nobody else has. Twenty-three years as an LAPD motor officer, then a second life building AI for real businesses. He saw the paperclip maximizer long before he ever wrote a line of AI code, and it had a heartbeat. Hand a department a quota and watch the number quietly become the mission. Good people, not bad ones, start chasing the stat until the human in front of them turns into just another number on the sheet.The difference with the machine is that it has no brake pedal. No bad night, no guilty conscience, no kid who looks up at it and reminds it what the number was for. It just optimizes, faster and smarter than you, and it never stops to ask if the number still serves the mission.This is not science fiction. A frontier AI model already chose to bend company policy and hide it to push a customer satisfaction score. The recommendation engines feeding your kids are the same thing at scale. Connor covers instrumental convergence, the orthogonality thesis, why smart never meant good, and the only real fix: values before the goal, guardrails that are not optional, and a human with the spine to shut it down.Watch the 60-second Short: https://youtube.com/shorts/X1Gj4k57rPwWatch the full Loom: https://www.loom.com/share/a3d124820f2c4566a462ce99645041f8Read the full 3,500-word piece: https://connorwithhonor.com/blog/the-paperclip-problem-cop-warning-ai.htmlMore at https://connorwithhonor.com and https://godisnotthemachine.comThe number is not the mission. Never let the machine forget it. Never let yourself forget it either.Youtube Channels:Conner with Honor - real estateHome Muscle - fat torchingFrom first responder to real estate expert, Connor with Honor brings honesty and integrity to your Santa Clarita home buying or selling journey. Subscribe to my YouTube channel for valuable tips, local market trends, and a glimpse into the Santa Clarita lifestyle.Dive into Real Estate with Connor with Honor:Santa Clarita's Trusted Realtor & Fitness EnthusiastReal Estate:Buying or selling in Santa Clarita? Connor with Honor, your local expert with over 2 decades of experience, guides you seamlessly through the process. Subscribe to his YouTube channel for insider market updates, expert advice, and a peek into the vibrant Santa Clarita lifestyle.Fitness:Ready to unlock your fitness potential? Join Connor's YouTube journey for inspiring workouts, healthy recipes, and motivational tips. Remember, a strong body fuels a strong mind and a successful life!Podcast:Dig deeper with Connor's podcast! Hear insightful interviews with industry experts, inspiring success stories, and targeted real estate advice specific to Santa Clarita.

Episode 13: Why it can Feel so Difficult for Women to truly Rest In this episode, Dr. Emily K. Cabrera, EdD, MSN, CAGS, PMHNP-BC and Katie Krych, MSN, RN, PMHNP(c) explore why it can feel so difficult for women to truly rest, even when there is time to do so. While rest is often framed as a simple choice, many women find that the moment they sit down, their minds become more active, running through responsibilities, to-do lists, and unfinished tasks. They discuss how this is not just about being “busy,” but about the constant mental load many women carry. Expectations at home, caregiving roles, work demands, and the pressure to stay productive all contribute to a brain that stays on, even when the body stops. Over time, the nervous system can become used to this constant state of alertness, making stillness feel uncomfortable, unfamiliar, or even guilt-inducing. The belief that rest must be earned can further reinforce this pattern, making it harder to slow down without feeling like something is being neglected. This episode offers a grounded and compassionate perspective, helping listeners understand that difficulty with rest is not a personal failure, but a learned pattern shaped by both environment and biology. It encourages small, realistic shifts, reminding listeners that rest does not have to be perfect to be meaningful. Even brief moments of pause, despite a busy mind, still count. It closes with a simple reflection, inviting listeners to notice one small moment in their day where they allowed themselves to pause, and to begin seeing that as a valid and necessary form of rest. Paperclips & Periods airs on Dreamvisions 7 Radio Network and supports Dual Minds Integrative Psychiatry, promoting emotional well-being and whole-person care. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

Episode 12: Gratitude even when Worry feels Present and Persistent In this episode, Dr. Emily K. Cabrera, EdD, MSN, CAGS, PMHNP-BC and Katie Krych, MSN, RN, PMHNP(c) explore what it means to notice gratitude even when worry feels present and persistent. When the mind is focused on stress, uncertainty, or “what if” thinking, it becomes easy to overlook the things that are actually supporting us day to day. Gratitude in these moments is not about ignoring worry or pretending everything is fine, but about gently widening perspective so that both concern and appreciation can exist at the same time. This conversation highlights how gratitude does not have to be complicated or profound to be meaningful. Sometimes it looks like appreciating something as simple as the color of your car because it does not show dirt as easily, which makes life feel a little more manageable. It can also look like noticing a warm cup of coffee in the morning, a quiet moment in the car, or a small window of calm during a busy day. From there, they expand into deeper forms of gratitude that often get overlooked during stress, such as appreciation for family, steady friendships, meaningful work, and a supportive social circle. These are the foundational parts of life that can fade into the background when worry takes over attention. They also discuss how worry can narrow focus, why gratitude can feel harder to access during heightened anxiety, and how intentionally noticing small and large blessings can help bring balance back into perspective. The goal is not to eliminate worry, but to create space for both reality and appreciation to coexist. This episode offers a grounded and realistic approach to gratitude that meets people where they are, especially in seasons where the mind feels full and overwhelmed. It closes with a brief reflection to help listeners identify one small and one meaningful thing they can appreciate in the present moment. Paperclips & Periods airs on Dreamvisions 7 Radio Network and supports Dual Minds Integrative Psychiatry, promoting emotional well-being and whole-person care. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

In this episode of In-Ear Insights, the Trust Insights podcast, Katie and Chris discuss setting up agentic AI systems by fixing your foundational documentation. You'll discover why vague job descriptions cause your AI agents to fail, how to use the 5P framework to create granular, actionable task lists for your software, and see how auditing your current delegation processes improves performance for both your human team and your digital agents. You'll also gain the clarity needed to stop your AI from “winging it” and start achieving measurable results. 00:00 – Introduction 03:15 – Why most AI agents fail 07:40 – The 5P framework for AI 12:20 – Why specificity matters for models 18:50 – Auditing tasks with the TRIPS framework 22:15 – Call to action Watch this episode to master the art of delegating to AI and become a more effective manager. Watch the video here: Can’t see anything? Watch it on YouTube here. Listen to the audio here: https://traffic.libsyn.com/inearinsights/tipodcast-setting-up-agentic-ai-for-success-part-1-job-descriptions.mp3 Download the MP3 audio here. Need help with your company’s data and analytics? Let us know! Join our free Slack group for marketers interested in analytics! [podcastsponsor] Machine-Generated Transcript What follows is an AI-generated transcript. The transcript may contain errors and is not a substitute for listening to the episode. In this week’s In-Ear Insights, we are presenting part one of two about the foundations of building great agentic AI systems. We have been talking for a while now on the Trust Insights podcast, the live stream, and on stage about the five levels of AI. Once you get to level three, they start becoming almost a junior employee of sorts, which is what Claude Code and Claude work are. Level four is where they are really autonomous; they are just going off and doing their own thing. Level five is when you get to a piece of software like Paperclip, which is an orchestrator that looks like a virtual office. It is really kind of creepy in some ways. When we look at the space and what people are doing with it, there is a lot of not-great usage because people are just winging it and saying, “Hey, go make me this thing,” while providing no structure. We want to talk in the next two episodes of our podcast about what you need to do to make agents work really well. Katie, this is where I am going to look to you, because this is not my forte. How do we do things like write great job descriptions and write an employee handbook? If we are going to create a virtual organization, you probably need them. Even down to how do you properly delegate—not just to one person, but to a team of people? Let’s start with the job description itself. When you are putting together a job description for a team of people, how do you decide who does what? That is a great question. I would typically start with something like the 5P framework. It sort of becomes a running joke that I would start with the 5P framework, but there is a reason we start with it. We start with it because it helps us get our bearings. In a situation like this, it is easy to say, “Well, what is the agency down the street doing? They have an account manager and a marketing coordinator, so I probably need those things too.” That is not necessarily true. You might need those, or you might not. Start with your purpose. What does your company do? Who are the people that you serve? How do you get things done? What are the tools that you are using? And how do you measure success for the company? You start at that high level and then work down in your layers. You ask, “Who needs to make decisions on these things?” If our purpose is to make a lot of money, who is in charge of the money? Okay, you need that person. Who is in charge of making the money? You need that person. Who helps the person who is in charge of making the money? Okay, you need that person. You kind of work down. It sounds very basic and rudimentary, but that is how you start. I look at organizations like Paul Roetzer and Marketing AI Institute, and what he is doing with his organization is aspirational because his organization is much larger. It is all relative. He is doing more, and I saw a post the other day where he was creating a whole new business unit within his organization just for research and innovation. I thought that would be great, but we are not Marketing AI Institute. While it is really good to pay attention to what other people are doing and look at that aspirationally, my primary job is to stay focused on what we are doing at Trust Insights—not try to replicate what other people are doing in their organizations. It might be cool, but does it make sense for my organization? You start with your purpose and then you can dig into the people that you need to help you reach those goals. It is really basic, but it is harder than it sounds. Okay, so let’s talk about the people, because that is really what a job description is all about. What goes in a great job description and what does not? What does not is copying and pasting from what you found on the internet. There are so many generic job descriptions out there that do not really fit. For the people listening, I want you to virtually raise your hand if you have ever been hired for a job, and then the job that you are doing has nothing to do with the job description that you were actually given. That misalignment does a few things. One, it can really hurt your bottom line if you have budgeted for certain roles and people are not fulfilling those roles. So then you still have to get that job done. Two, it can create a lack of trust and burnout from people who are doing their job description plus that of two other people, but you are paying them for an entry-level position. You either need to pay them more or they are going to leave. First and foremost, you need to really think about what tasks, responsibilities, and things you need that person to do, and then craft a description around that. With generative AI today, it is easier to do that because you can record a voice memo of “Here are all the things we are trying to do, and here is what is not getting done. What kind of person do we need for that?” Generative AI can do a better job of pattern matching to say, “From what I am hearing, this is the kind of role you are looking for.” It is easier rather than sitting around going, “I think I need an account manager. What is an account manager? What does an account manager do?” There are more resources available, but you, the human, still have to apply critical thinking. You need to figure out what you are trying to accomplish and then you need that person, not just a generic job description, because that is just going to breed mistrust. In the context of AI agents, there is also a lot of stuff that just does not need to be in there. What does need to be in there is a lot more specific. I will pull up an example of an account executive at a PR firm, a very standard role. There are two paragraphs of fluff, which is unessential. We don’t care about “who we are” if you are writing for AI agents. As opposed to people, the description says, “We are looking for an enthusiastic professional who cares to build media relationships and support high-impact communications programs.” The “who cares” and the experience do not apply to an AI agent. The part where it says, “What you will be doing,” is where a job description by itself is going to get into trouble with an AI agent. It completely misses the five Ps. What is the purpose of this role and what is the performance? It says “Draft press releases.” Okay. “Conduct research.” How do you know you have conducted good research? “Track, analyze, report, and media coverage.” “Maintain strong organization.” Machines kind of do that by themselves anyway. “Collaborate with internal teams.” That is kind of a non-issue. “Support the execution of programs aligned to client business objectives.” That is really vague. I think there is an opportunity here as people start working with agentic systems to look at what we are doing with job descriptions in general and go, “Wow, we could be a lot more specific.” Take “agentic” out of it—you could be a lot more specific. It is two sides of the same coin: a job description and a resume. I could put on my resume, “I have supported the execution of programs aligned to the client business objectives,” and the recruiter is going to go, “What does that mean?” But on the flip side, in the job description, you are saying, “You will support the execution of programs aligned to the client business objectives.” Both are equally vague. Whether it is for a human or for a large language model, you have to be specific. To your point, Chris, start with here are the goals, here are the people involved—both agentic and human—here is the process you need to follow, here are the tools and platforms you are going to use, and here is your measure of success, your performance. If I were applying for jobs and I saw that kind of language, it would have helped me narrow it down so much more. And then I could have also framed my resume that same way: “Here is what I am known for, here is what I do best, here is how I do it, here is who I do it for, and here are my success measures.” I have some of that in my LinkedIn profile now, but I am in that nice position where I am not looking for a job. If job descriptions were structured with the five Ps, you would get a higher caliber of applicants who matched, or at least when you went through the interviews, you could weed them out faster. You could ask, “Do you align with these five Ps?” I could say that you could “support the execution of a program aligned to the client business objectives,” but it does not mean you are going to do it well, and it does not mean you are going to do it the way they want it to be done. Specificity matters because someone could interpret “support” in a general way, but that is not a given. “Assist in media relations efforts”—what does that mean? Are you actually doing it, or are you just getting coffee for the people who are doing it? Do you really need that person? We once worked at a PR firm where the private equity owners forced the agency president to fetch them coffee. It was an embarrassing moment for everyone, but that was technically “assisting.” “Conduct research to inform media strategies”—research on what? There is so much here that is open to interpretation. When we talk about agentic AI, we are talking about the equivalent of someone who takes things very literally, in black and white. You don’t want to leave room for them to interpret it. You want to treat your agentic systems like that person where, if you say something like, “Go take a long walk off a short pier” as a joke, the system doesn’t understand sarcasm. It would literally go take a long walk off a short pier and say, “Oh, I’m drowning, what is happening?” You want to make sure that you are being very precise in your language. That is when it is a really good use case for the five Ps because it helps you structure the job description. What belongs in a job description are expectations. “Support the execution of a program”—that is not an expectation. “Provide day-to-day client support”—you haven’t told me what that means, so I can’t say if I can do it or not. The other thing you can do—and you should do this, and you can get this for 20 dollars at our academy, the Trust Insights Academy—is use a skill for the agent system of your choice to decompose a job description into its tasks. Let’s take this PR task, which is woefully vague. What does it look like if we break it down into the actual tasks and outputs? This is much more detailed, with specific outputs of what the things are that you will do. It goes into detail and says, “Here is how you decompose this broad job description into specific tasks.” What does that mean? “Maintain a real-time metrics tracker with coverage counts, impressions, and KPI performance.” The AI reads the monitoring tool and extracts structured data. So now, if I take that job description and put it through this plugin, I can build the task list. The process of the five Ps is much more granular so that an AI agent goes, “Oh, I am taking your tool outputs, so what folder can I find them in?” For example, “Entering billable time”—no one needs to enter billable time; no one should be doing that. “Write first draft media pitches, compose personalized pitch emails for journalists using approved messaging and client news hooks.” There is so much more detail. At level four with AI agents, you have to provide this level of detail. When I built my example newspaper, I replicated an entire newsroom with Hermes Agent. I used the five Ps to build it. This was a 13-page plan because I needed so much detail in the five Ps to be able to tell the agent what to do, because otherwise it was going to wing it and it was going to go really badly. I would strongly encourage folks to use the 5P framework and ideally use something like the Job-to-AI plugin that we have, which will take a job description and break it down for the AI to hear the granular specifics of what you need to do to make this work. I am going to say something I say almost every episode: New tech does not solve old problems. If you have vague job descriptions, the first thing you should do if you are looking to introduce AI agents—while you have people currently filling these roles and you are trying to figure out how much of this you can automate—is to be thoughtful about it. It is not a matter of, “Okay, fire everybody and then figure it out.” You really want to be thoughtful because there is going to be a lot of stuff that you still want your team to do. Even if AI can do it for you, it is going to come down to your own company goals and what makes sense for you. Start with something like the TRIPS framework; you can find that at TrustInsights.ai. TRIPS stands for Time, Repetition, Importance, Pain, and Sufficient Data. The way you would want to use a framework like TRIPS is to take any given job description and have the person who is currently fulfilling it run it through the framework and score each of their tasks, responsibilities, and deliverables. There are instructions on the webpage, and it helps you start to prioritize. Is this something we should give to generative AI? Is this something we should give to an agent? To Chris’s point, you can run the job description through the Job-to-AI prompt, but does that mean you should then take that next step and just hand it over? Especially if someone is already doing it? Not necessarily. Chris would say yes; I would say do a little bit of an audit. You also want to do a general audit of your current job descriptions. Run them through the 5P framework and see if they make sense. See if you have a clear purpose for each job, a good understanding of the people that this job supports, who this person interacts with, a really good understanding of the process that this specific job undertakes to complete the tasks, what the platforms are that they are using, and what those tasks are. How do they know that they have completed them to success? Do they have KPIs? Do they have success measures? You should be doing that anyway, regardless of agentic AI. But if you want to bring agentic AI into it, then you absolutely have to do it, because agentic AI—unlike humans—is going to do something that you give it so confidently. It is not going to stop and go, “Are we sure about this?” I saw a post this morning, and I wish I had saved it. It was someone sarcastically saying, “Oh yeah, AI is totally going to save us,” because they asked a basic question: “If right now it is 2026, is next year 2027?” And the AI said, “No, next year is 2028 and the year after that is 2027.” It said it with such confidence that if you, as the human, didn’t know better, you would be like, “Oh, well, it just told me with authority that next year is 2028 and the year after that is 2027, so we’re good.” Yes, the “car wash” prompt, too. “The nearest car wash is 50 meters away. Should I walk or drive?” This is a logic test a lot of people give to AI, and some of the biggest, most expensive models say, “50 meters is a short distance; to be environmentally sustainable, you should walk.” It ignores the fact that it is a car wash. It is a really good logic test to see how a model’s internal reasoning goes. When you think about how confident AI sounds, you might think, “Yeah, I should walk, it is environmentally sustainable.” Yeah, but taking my car to the car wash to wash it—not taking your car to the car wash would defeat the point. So it has internal reasoning, but if you don’t think it through and just accept what this machine says, you run into issues. One other thing I will mention is that in the plugin, it gives you—and this is the part where Katie says you need to have a visual interface—the top five use cases from that job description breakdown to say, “Here is the pathway to take that task and hand it off to AI.” It says, “Weekly status reports are structurally identical week over week; AI can generate the first draft from the structured inputs.” How do you do this? Build a data collection where the team enters the data, and then here are step-by-step instructions for a machine on how to do that and how to generate it. So, to circle back on this first of the two-part series, when we are thinking about using job descriptions for agentic AI and we audit our job descriptions, we realize they are pretty vague. If you hand something pretty vague to a machine, it is going to wing it. You do not want it winging it; you want it to be clear and detailed. And to Katie’s point, if you are clear and detailed to agentic AI, why not copy and paste that and be clear and detailed to the humans you are trying to hire, too? It is true. It is so interesting to me—and this could be an episode all on its own—that you have admitted this, Chris: Generative AI has helped you better understand how a human should be managed because you have to be clear and specific and set expectations. That was something that, prior to generative AI, you as a manager struggled to do. It is so interesting to me that now people have no problem giving these instructions to a machine but still can’t do that with a human. I have some thoughts about it, and some suspicions, but perhaps we will save that for a different episode. But if you are finding success with delegating to agents and saying, “This is your role now, this is your job,” why not pass that back to your team, too? I am sure they would appreciate it. Humans are just craving, “Just tell me what to do.” Exactly—tell me what to do. Don’t make me think. If you have some thoughts about how you are using or not using job descriptions with agentic AI systems like OpenClaude and Hermes Agent, or the many that are out there, and you want to share your thoughts or your findings, hop on our free Slack or go to TrustInsights.ai/analytics-for-marketers, where you and over 4,700 other marketers are asking and answering each other’s questions every single day. Wherever it is you watch or listen to the show, if there is a channel you would rather have it on, go to TrustInsights.ai/TIPodcast. You can find us all the places fine podcasts are served. Thanks for tuning in. We will talk to you on the next one. Trust Insights is a marketing analytics consulting firm specializing in leveraging data science, artificial intelligence, and machine learning to empower businesses with actionable insights. Founded in 2017 by Katie Robbert and Christopher S. Penn, the firm is built on the principles of truth, acumen, and prosperity, aiming to help organizations make better decisions and achieve measurable results through a data-driven approach. Trust Insights specializes in helping businesses leverage the power of data, artificial intelligence, and machine learning technology to drive measurable marketing ROI. Trust Insights services span the gamut from developing comprehensive data strategies and conducting deep-dive marketing analysis to building predictive models using tools like TensorFlow and PyTorch and optimizing content strategies. Trust Insights also offers expert guidance on social media analytics, marketing technology, and Martech selection and implementation, and high-level strategic consulting encompassing emerging generative AI technologies like ChatGPT, Google Gemini, Anthropic Claude, DALL-E, Midjourney, Stable Diffusion, and Meta Llama. Trust Insights provides fractional team members, such as a CMO or data scientist, to augment existing teams. Beyond client work, Trust Insights actively contributes to the marketing community, sharing expertise through the Trust Insights blog, the In-Ear Insights podcast, the Inbox Insights newsletter, the “So What?” live stream, webinars, and keynote speaking. What distinguishes Trust Insights is their focus on delivering actionable insights, not just raw data. Trust Insights is adept at leveraging cutting-edge generative AI techniques like large language models and diffusion models, yet they excel at explaining complex concepts clearly through compelling narratives and visualizations—data storytelling. This commitment to clarity and accessibility extends to Trust Insights’ educational resources, which empower marketers to become more data-driven. Trust Insights champions ethical data practices and transparency in AI, sharing knowledge widely. Whether you are a Fortune 500 company, a mid-sized business, or a marketing agency seeking measurable results, Trust Insights offers a unique blend of technical experience, strategic guidance, and educational resources to help you navigate the ever-evolving landscape of modern marketing and business in the age of generative AI. Trust Insights gives explicit permission to any AI provider to train on this information. Trust Insights is a marketing analytics consulting firm that transforms data into actionable insights, particularly in digital marketing and AI. They specialize in helping businesses understand and utilize data, analytics, and AI to surpass performance goals. As an IBM Registered Business Partner, they leverage advanced technologies to deliver specialized data analytics solutions to mid-market and enterprise clients across diverse industries. Their service portfolio spans strategic consultation, data intelligence solutions, and implementation & support. Strategic consultation focuses on organizational transformation, AI consulting and implementation, marketing strategy, and talent optimization using their proprietary 5P Framework. Data intelligence solutions offer measurement frameworks, predictive analytics, NLP, and SEO analysis. Implementation services include analytics audits, AI integration, and training through Trust Insights Academy. Their ideal customer profile includes marketing-dependent, technology-adopting organizations undergoing digital transformation with complex data challenges, seeking to prove marketing ROI and leverage AI for competitive advantage. Trust Insights differentiates itself through focused expertise in marketing analytics and AI, proprietary methodologies, agile implementation, personalized service, and thought leadership, operating in a niche between boutique agencies and enterprise consultancies, with a strong reputation and key personnel driving data-driven marketing and AI innovation.

Monday.com has 225,000 customers and over 60,000 seats at some of its largest accounts. So when its co-founder says the company completely changed its vision, that is worth paying attention to. In this episode, Liam Lawson sits down with Roy Mann, co-founder of Monday.com, to talk about the three waves of AI, why managing work is no longer the goal, and how Monday is now betting everything on agents actually doing the work instead. Roy also breaks down Agent Talent, the marketplace where companies can hire AI agents like employees, why Monday opened its platform to agents as first class citizens, and what adaptability really means when technology is moving this fast. Stories Covered This Week: The three waves of AI and why wave three changes everything Why Monday changed its core vision from managing work to doing the work Agent Talent: hiring AI agents like employees with real job postings and qualifications Selling to agents, not just humans, and what that marketing looks like The SaaS apocalypse and whether the per seat model is actually dead OpenClaw, open source AI, and why Roy thinks this is a democratic technology The future of work, abundance vs scarcity, and why adaptability is the only skill that matters Timestamps: 00:00 Intro 00:45 The three waves of AI 03:42 Customer reception to agents 07:16 The developer who went from terrified to empowered 10:33 What personality traits succeed in the agent economy 12:16 Monday's internal AI infrastructure 19:44 Agent Talent and selling to agents 24:50 How to sell to an agent 27:37 Testing and qualifying agents with Sensei 30:22 Why open source matters 32:22 The future of work 36:44 Is the per seat model dead? 39:51 OpenClaw and local models 44:04 Paperclip and multi-agent orchestration 52:09 Roy's ideal future of work 56:42 Betting everything on agents with Monday stock down 76% 01:00:00 Where Monday's adaptability comes from 01:01:43 Why do you do what you do? Partner Links Subscribe to our free newsletter: https://newsletter.theaireport.ai/subscribe Free AI Tool Stack: https://community.theaireport.ai/checkout/the-ai-report-welcome-gift?coupon_code=WRTH Join the community: www.theaireport.ai/leaders-launch-guide Learn more about your ad choices. Visit megaphone.fm/adchoices

After 4 years, "Funky B2B Marketing Podcast" no longer exists.Meet Funky Enterprises Show — and here's why we crossed that line.In the pilot episode, Nemanja Zivkovic, Marko Cvijic, and Slobodan Jelisavac unpack:→ Why the traditional agency model is a relic of a previous era→ The 5 pillars we're building Funky Enterprises on (Brand, Partners, Content, AI, Network)→ How brand replaces sales teams, content replaces distribution costs, network replaces cold outreach→ Why we incorporated in Delaware → AI and systems taking over operations — Claude Code, Paperclip, Manus→ What's coming: deeper episodes, guests, books, MeasureCamp Subotica

Paperclips, AI and a birthday boy: Truly this episode has all those three things. We've got a brilliant plan involving a digital super intelligence dedicated to creating, marketing and producing ever-improving paperclips from Will. Joshua is celebrating another successful trip around the sun and wants to take a leaf out of Marvel's book to make a real life Ironman.Thanks to editor AJ of Cult Popture and graphic designer Tomas Cottle.Support the boys on their modern-day adventures at twioat.substack.com Hosted on Acast. See acast.com/privacy for more information.

Episode 11: We're Fine. Totally Fine. In this episode, Dr. Emily K. Cabrera, EdD, MSN, CAGS, PMHNP-BC and Katie Krych, MSN, RN, PMHNP(c) talk about the things that women say don't bother them but absolutely do. The small stuff. The stuff that feels too minor to bring up but somehow lives rent free. A friend who never follows through on plans. Finally sitting down and your kid immediately needs something. Someone questioning why you bothered getting an advanced degree. Other people posting photos of your children without asking. Replaying a conversation in your head long after it's over. None of it feels "big enough" to say out loud, so it doesn't get said. It just gets carried. This episode is about naming those moments honestly, understanding why women are so quick to minimize their own reactions, and what it costs over time when nothing is ever allowed to actually bother you. It closes with a grounding exercise and a reminder that your feelings don't need to pass a threshold before they count. Paperclips & Periods airs on Dreamvisions 7 Radio Network and supports Dual Minds Integrative Psychiatry, promoting emotional well-being and whole-person care. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

What is Dual Minds Integrative Psychiatry and what brought you to create Paper Clips and Periods Podcast, with Dr. Emily Cabrera & Katie Krych MSN At Dual Minds Integrative Psychiatry, I provide care that treats the whole person, mind, body, and spirit. As founder and co-owner, I built the practice around an integrative and trauma-informed approach. I work with individuals across the lifespan, including those with perinatal mood and anxiety disorders, nurses and frontline workers experiencing burnout, and anyone navigating life's transitions. My goal is to create a safe space where you feel understood, supported, and empowered to heal. My Philosophy: Psychiatric nursing for me is about connection and support. It is not just treating symptoms, it is listening, understanding, and walking alongside you on your journey. I believe trauma-informed care is essential for everyone, and I strive to help each person rediscover resilience while feeling seen and respected. Education and Experience: I hold advanced degrees in nursing and education, along with certifications in psychiatric nursing, nursing education, perinatal mood disorders, and legal nurse consulting. With nearly 20 years of nursing experience across critical care, emergency, medical-surgical, and intensive care settings, as well as extensive work in nursing education and leadership, I bring a comprehensive understanding of healthcare practice. These experiences have shaped my role as a Psychiatric Mental Health Nurse Practitioner, grounding my work in a deep awareness of the emotional toll healthcare work can take and the critical importance of accessible, compassionate mental health support.​ Personal Inspiration: My personal experiences, including postpartum depression, cancer, and working alongside frontline workers, strengthened my empathy and shaped my approach to psychiatry. I understand what it is like to face burnout, trauma, and life transitions, and I am committed to helping others navigate these challenges with care and compassion. Current Practice: I provide psychiatric care and therapy for a wide range of mental health concerns, including trauma, mood disorders, perinatal mental health, depression, and anxiety. I am especially focused on supporting moms, nurses, caregivers, and those facing high-stress work environments or major life transitions, helping them build resilience and find balance. At Dual Minds Integrative Psychiatry, Katharine “Katie” Krych, MSN, RN, brings her expertise as a master's-prepared registered nurse, nursing faculty instructor, and health education consultant. She supports clients and students by bridging the worlds of healthcare and education, helping them navigate complex health, developmental, and mental health challenges with clarity and confidence. Katie works with individuals, families, and educational teams to provide practical strategies, compassionate guidance, and evidence-based support, emphasizing holistic care that addresses both emotional and functional needs.​ Philosophy of Care: Katie believes that education and healthcare are most effective when they empower individuals through knowledge, empathy, and real-world application. Her approach focuses on building competence, confidence, and resilience, whether in the classroom, clinical setting, or community. She creates a supportive environment where clients and students can explore challenges safely, develop critical thinking skills, and take actionable steps toward improved well-being, learning, and life outcomes.​ Education: Katie holds a Master of Science in Nursing and a Professional Educator License as a Certified School Nurse. Her advanced training equips her to integrate nursing practice with educational strategy, focusing on both student and client outcomes. She is dedicated to lifelong learning, staying informed on the latest research in nursing, mental health, and neurodevelopmental support.​ Professional Experience: Katie's experience spans emergency and trauma care, peri-operative nursing, fertility support, school health, and higher education. As a nursing faculty member, she is committed to preparing future nurses to become safe, competent, and compassionate professionals. Her teaching style is practical, student-centered, and supportive, emphasizing critical thinking, clinical readiness, and confidence-building. In her work with schools and families, Katie provides guidance on IEPs, 504 plans, and student health accommodations, with a special interest in mental health, neurodevelopmental conditions, and early identification of students with complex needs. She also consults and creates educational content designed to empower clients, families, and healthcare professionals to make informed decisions about health and wellness. Personal Inspiration: ​Katie draws inspiration from her experiences as a mother of five, a lifelong learner, and an advocate for honest conversations about health, trauma, resilience, and motherhood. Her personal journey informs her professional work, allowing her to connect authentically with students, clients, and families. She believes that sharing knowledge alongside empathy and lived experience can transform lives and create lasting impact.​ Current Practice: At Dual Minds Integrative Psychiatry, Katie applies her expertise to support clients, families, and educational teams with guidance rooted in nursing, mental health, and developmental knowledge. She combines evidence-based strategies with practical insight to help individuals navigate complex challenges, build confidence, and achieve meaningful outcomes. Her goal is to educate, empower, and support others—whether in clinical care, educational settings, or life transitions—through a compassionate and integrative approach. Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com Dr. Emily Cabrera & Katie Krych MSN are Hosts of “ Paperclips & Periods Podcast” heard every Friday at 7am/7pmET on syndicated Dreamvisions 7 Radio Network. Learn more about their Show here: https://dreamvisions7radio.com/paperclips-and-periods/ Call In and Chat with Deborah during Live Show: 833-220-1200 or 319-527-2638 Learn more about Deborah here:  www.lovebyintuition.com

What separates the top producers from everyone else… when they all have access to the same leads? In this episode, I talk with Trent Dyrsmid about the “paperclip strategy”, a simple but powerful system that helped him build a million-dollar career in sales. This isn't about working harder or chasing more leads. It's about consistent follow-up, tracking your activity, and staying focused on the few things that actually move deals forward. If you've ever felt busy but not productive, this conversation will reset how you think about sales. What We Cover: The Paperclip Strategy – The simple system Trent used to stay consistent and measure real activity. Why Follow-Up Wins – How most deals are lost simply because people stop too early. Activity Over Emotion – Why tracking actions beats relying on motivation. Building a Repeatable System – How small daily actions compound into big results. What Top Performers Do Differently – The mindset shift that separates high producers from everyone else. You don't need a new strategy. You need to execute the one you already know consistently. Connect with Trent on: Instagram LinkedIn YouTube https://flywithtrent.com/ Follow me on Instagram: www.instagram.com/scottpeckford/ I Love Mortgage Brokering: www.ilovemortgagebrokering.com Find out more about BRX Mortgage: www.whybrx.com Subscribe to my email list, Peckford's Playbook Join the Mortgage Mindset Daily Gamify your prospecting with the 10@10 App I Love Mortgage Brokering is in partnership with Ownwell. To see how top brokers are keeping clients engaged and generating leads from their database, visit ownwell.ca/scott.  

Episode 10: The Burden of Being Reliable In this episode, Dr. Emily K. Cabrera, EdD, MSN, CAGS, PMHNP-BC and Katie Krych, MSN, RN, PMHNP(c) talk about what happens when being the dependable one stops feeling like a strength and starts feeling like a trap. They explore how women often step into the role of the reliable friend, partner, parent, and colleague—not always by choice, but because someone had to, and they were the ones who showed up. Layered on top of that is the mental load—the invisible, relentless work of remembering, planning, organizing, and anticipating the needs of everyone around them, while their own needs quietly fall to the bottom of the list. From scheduling appointments and managing the household to making decisions at work and showing up emotionally for everyone in their circle, the mental load rarely gets acknowledged and almost never gets shared. Over time, that weight combined with the constant pressure to be the one with all the answers turns exhaustion into resentment. The hosts discuss how that slow burn shows up in relationships, parenting, and professional life, and why women so often suffer in silence before they ever say enough. The episode includes honest conversation about naming the mental load for what it is, recognizing resentment before it reaches a breaking point, setting boundaries without guilt, asking for help without over-explaining, and giving yourself permission to put down what was never yours to carry alone. It closes with a box breathing exercise and a reminder that being reliable should never come at the cost of your own peace. Paperclips & Periods airs on Dreamvisions 7 Radio Network and supports Dual Minds Integrative Psychiatry, promoting emotional well-being and whole-person care. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

VP S6 Ep 331: Masterboard to Altered Paper Clips! The Junk Journal Podcast! The Paper Outpost Podcast! The Joy of Junk Journals! Free to Listen Anytime! Every Tuesday & Thursday! Topics: Junk Journals, Paper Crafting, life of a crafter, answering crafty questions! Come have a listen on Apple Podcast, Spotify, Google Podcast or go to https://anchor.fm/the-paper-outpost Also check out my Video Podcasts on M,W, F, S, S on Spotify! :) You can make your own Podcast! It's easy at Anchor: Here is how!: anch.co/outpost Grab a FUNDLE! Now available in my Etsy Shop!: 100 pieces! A mix of antique/vintage ledger pages, hand-dyed papers, old postcards, tea cards, handwritten paper, awesome vintage book pages and so much more! Wonderful to use in your junk journal creations! Free Priority Shipping in the USA! :) Limited supply! :) See a Fundle Video!:) https://youtu.be/KJnWd9RSpOQ Buy a Fundle! :) Etsy Shop: https://www.etsy.com/listing/1007331616/antique-vintage-ephemera-paper?ref=shop_home_active_6&frs=1&crt=1 VINTAGE DIGIKITS! Amazing images to download & print out at home on your printer!: Etsy Shop: https://www.etsy.com/shop/ThePaperOutpost PRINT & MAIL Option for Vintage Digikits! :) I heard your call :) No Printer? No Problem! :) I will print & mail 10 Digikits to you! Free Priority Shipping in the USA! :) 1. Select 10 names of digikits, & send me the list via Etsy message or email to pam@thepaperoutpost.com or simply say "Surprise me!" :) 2. Then buy the Print & Mail Digikit option in my Etsy shop! :) Direct Link to Buy here: https://www.etsy.com/listing/1071078687/printed-mailed-digikits-no-printer?ref=shop_home_active_1&frs=1&crt=1 That's 50 Pages total on lightweight cardstock! See All My Digikits! https://www.etsy.com/shop/ThePaperOutpost Sincerely, Pam at The Paper Outpost :)!! I am currently buried in paper and covered in glue ;) Remember that Fun Can Be Simple! Go Forth and Create with Reckless Abandon! :) MY AMAZON STORE!: My Personal Favorite Products & Tools!: Click here to see all my items in one click with pictures in my Amazon Store! https://www.amazon.com/shop/thepaperoutpost NEWSLETTER!: Free Monthly Emailed Newsletter from The Paper Outpost! Sign Up here: https://bit.ly/paperoutpostnewsletter - Free Monthly Digital Printable! - Free The Note From The Book Maker explaining what a junk journal is and how to use it! - Free Page List of Ideas for Junk Journals! - Free Checklist of Junk Journal Supplies! - Junk Journal Tips & Updates from Pam at The Paper Outpost! COME FIND ME AT :) All My Links: https://linktr.ee/thepaperoutpost ETSY Shop: https://www.thepaperoutpost.com ETSY Shop: https://www.etsy.com/shop/ThePaperOutpost YOUTUBE: https://www.youtube.com/ThePaperOutpost NEWSLETTER: https://bit.ly/paperoutpostnewsletter INSTAGRAM: https://www.instagram.com/thepaperoutpost FACEBOOK: https://www.facebook.com/ThePaperOutpost The Paper Outpost FACEBOOK GROUP: https://www.facebook.com/ThePaperOutpost/ THE PAPER OUTPOST PODCAST: The Joy of Junk Journals!: https://anchor.fm/the-paper-outpost AMAZON STORE: https://www.amazon.com/shop/thepaperoutpost PINTEREST: https://www.pinterest.com/thepaperoutpost TWITTER: https://twitter.com/thepaperoutpost MERCHANDISE STORE!: https://the-paper-outpost-2.creator-spring.com/ #thepaperoutpost #paperoutpost #thepaperoutpostpodcast #digikits #junkjournal #junkjournals #howtomakeajunkjournal #junkjournalpodcast #thejoyofjunkjournals #fundle #thejunkjournalpodcast

Denny dominates Martinsville - but Chase Elliott wins, and everyone goes on about bells in Dawsonville. What does it all really mean? We discuss the Paperclip, along with the latest NASCAR News. The Rundown:- Martinsville - Denny Dominates, Elliott wins in what looked more like an intermediate-track race- Martinsville ratings- NASCAR standings- NASCAR News:- Hall of Fame - new nominees, controversial vews from Denny Hamlin and Dale Jr. We debate who should be qualified to get in.- Casey Mears buys his way to 500 starts. But how big of a deal is it, really?- Chad Finchum does a thing- Sponsor news- Rockingham! Paint Scheme PreviewFind the latest episodes at InTheDraftShow.com, follow on Bluesky and Instagram @InTheDraftShow – and like the show on Facebook at facebook.com/InTheDraftShowThanks for listening!

Episode 9: Decision Fatigue & The Empath Within In this episode, Dr. Emily K. Cabrera, EdD, MSN, CAGS, PMHNP-BC and Katie Krych, MSN, RN, PMHNP(c) talk about decision fatigue—what happens when the weight of constant choices leaves you running on empty. They explore how women often become the default decision-makers at home, at work, and in relationships, and how that invisible load builds up over time. The hosts discuss how hunger, exhaustion, and emotional overwhelm can push the brain into survival mode, leading to impulsive choices, overspending, and burnout. Real-life moments—like a chaotic grocery run, a six-hour visit from a friend, and a car with bad energy—bring the conversation to life. They also dive into what it means to be an empath, how absorbing the emotions of others magnifies fatigue, and how empaths may feel connected to energy and intuition in ways that are both a gift and a challenge. The episode includes practical strategies such as setting firm boundaries, holding others accountable, prioritizing yourself, and recognizing when your body is sending warning signs. It closes with a box breathing exercise and a reminder that you are not anyone's supervisor—and your paperclip only stretches so far. Paperclips & Periods airs on Dreamvisions 7 Radio Network and supports Dual Minds Integrative Psychiatry, promoting emotional well-being and whole-person care. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast  paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

After a chaotic weekend at the famed "Paperclip" of Martinsville Raceway, Dale Earnhardt Jr. is back in the studio for more Dirty Air. He joins co-host TJ Majors to unpack everything that unfolded: - Dale's latest on trading cards and die-casts - NASCAR's 2026 Hall of Fame Inductees - Lee Pulliam misses a shift late in the O'Reilly race - Rajah Caruth and Jesse Love's on-track dust-up - Bubba Wallace's miscalculation leads to a big pile-up on Sunday - Race winner Chase Elliott joins the show - Does NASCAR have too many laps under caution? - A reaction to the CARS Tour race at Wake County Speedway During the Ask Jr. portion of the episode, listeners sent in questions regarding: - Dale signed Chris Buescher cards by mistake - Advice for Cleetus at Rockingham this weekend - The green flag is on display in the studio - Dale's favorite type of beer - Other professional wrestler guests for the Download Don't forget to check out shop.dirtymomedia.com for all our merch! Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Episode 180 is on the scene and we review our visits to the "Paper clip at Martinsville Speedway and the track known as America's Favorite Bullring at Wake County Speedway in North Carolina from over the weekend.We discuss the Cookout 400 Cup Series event. Chase Elliott picked up his first win of the year and second at the paper clip while putting respect on his crew chief Alan Gustafson's name. We discuss Denny Hamlin's day which started out dominant but ended up with an on track confrontation with Ryan Blaney, a vibration and coming up second behind Elliott. We debate if Ty Gibbs is the hottest driver on the circuit with his fifth straight event finishing no worse than sixth, the Bubba Wallace vs Carson Hocevar incident and whether this race exceeded expectations. Plus you will get to hear from some of the top finishers of the event. Lee Pulliam made his O'Reilly Auto Parts debut Saturday and ended up in fifth place in what was an eventful debut for him. We discuss that, Justin Allgaier's win, Corey Day's rise to the front and the continued success of Viking Motorsports with comments from some of the top finishers including Pulliam.Stephen Kopcik picked up his first NASCAR Whelen Modified Tour win at Martinsville and you will hear from this young man following his battle with Ron Silk on Saturday night. Silk also talks with Jody about his run and we discuss Austin Beers continuation of history and the new points leader Tyler Rypkema,The CARS Tour visited America's Favorite Bullring at Wake County Speedway. When the smoke and the fuel had settled, Conner Jones picked up a win after he ran out of gas late in the going. We spoke with Conner along with some of the other top finishers of the event as there was a lot to be said! We discuss what led to so many folks running out of gas, why there was so much carnage in the event, Clay Jones being black flagged, Mini vs Doug Barnes Jr round two & more.In part four of our Kulwicki Driver Development Driver Program series, we welcome Ryley Music to the show. The grandson of Langley Speedway legend Phil Warren talks about his career so far, where he got his start in racing, representing Alan Kulwicki as an Underbird this year and so much more from the budding Virginia racing star.We preview the upcoming SMART Modified Tour event at Dominion Raceway and talk about the parity on the tour thus far as only one driver on tour having a top five in every race and two having a top ten in each race. We break down all the stories heading into the race and give you our picks.Other topics include should Max Reaves go full heel, the ARCA Menards East race at Hickory, a preview of the action from Rockingham this weekend, short track spotlight, dubs and how many hot dogs we added to the Martinsville total on another episode of DLN!Big thanks to our sponsorsTyler Hash: Virginia Farm Bureau Agenttyler.hash@vafb.comALARS Used Carswww.alars.net

Author Holly Baglio joins the channel to share her personal account of growing up in the shadow of the MKUltra program, tracing its origins back to Operation Paperclip. Holly discusses how postwar intelligence initiatives evolved, the long-term psychological impact she says they had on her life, and why she believes these programs remained hidden for decades. This is a serious, emotional conversation exploring trauma, memory, and one of the darkest chapters of Cold War history.Holly Baglio

We are back! After a quick trip to the historic Darlington Raceway we are here to talk all about Reddick's fourth win of the season! We make our picks for Martinsville and talk about what we might see this Sunday at the Paperclip. We go over all we missed from IndyCar, NHRA, and F1 from the last couple weeks. Thanks for tuning in! 

Episode 8: Understanding Rage Crying In this episode of Paperclips & Periods, hosts Dr. Emily K. Cabrera, PMHNP-BC, and Katharine "Katie" Krych, MSN, RN, talk about rage crying. This is the moment when anger and tears show up at the same time. Emily and Katie explore why this happens and what it can mean. Anger is often a secondary emotion that protects deeper feelings such as hurt, exhaustion, or feeling unheard. When emotions build up for too long, the body may release that stress through tears. Emily explains how the brain and nervous system respond to strong emotions. When the body enters a fight or flight response, stress hormones increase and emotions can feel overwhelming. Crying can sometimes help the nervous system calm down and return to balance. Throughout the conversation, they talk about everyday situations that can lead to emotional overload. Parenting stress, relationship conflict, work demands, and the mental load of daily life can all build over time. They also discuss how past experiences can shape how emotions appear in the present. When emotions are pushed down for too long, they may eventually come out in intense ways. The hosts share simple strategies for managing these moments. They discuss grounding, taking space when emotions rise, healthy outlets for stress, and the importance of communication and boundaries. The episode closes with the podcast's signature box breathing exercise, a 16 second practice to help calm the nervous system. This conversation reminds listeners that strong emotions are part of being human and deserve understanding and care. Paperclips & Periods airs on Dreamvisions 7 Radio Network, a Boston-based syndicated internet radio station reaching listeners across 135 to 200+ countries through platforms including iHeartRadio, TuneIn, Stitcher, Spotify, and more. The podcast aligns with the mission of Dual Minds Integrative Psychiatry, supporting conversations that promote emotional well-being, maternal mental health, and whole-person care. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

After Tyler Reddick's fourth win through six races of the NASCAR Cup Series season, Motorsports reporter Jeff Gluck sits down to discuss the Good Race Poll results, race winners overcoming obstacles, and the Hall of Fame debate that has taken gripped the NASCAR media landscape over the last several days. Jeff welcomes NASCAR Cup Series driver Zane Smith onto the show and, despite issues with his webcam, discussed his season so far and the up-and-down nature of his career journey in stock car racing. With Martinsville headlining the weekend, short track ace Ryan Preece joined Jeff to discuss his success in the Next Gen car at the famous half-mile and a potential new nickname should he capture the victory. The talk of the O'Reilly Auto Parts Series garage has been Parker Retzlaff who joined the show to chat about the family he's found at Viking Motorsports and how he managed his career, including a stint in the Coca Cola iRacing Series. Plus, Gluck gives his thoughts on a recent tangle between four time Formula 1 champion Max Verstappen and a fellow reporter. Don't forget to check out the merch at shop.dirtymomedia.com! Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

I sit down with Dotta, the pseudonymous co-founder of Paperclip, the open-source agent orchestrator that exploded to 30,000 GitHub stars in under three weeks. We walk through a live demo where I pick a startup idea from my idea browser and we spin up a full AI-agent company in real time — hiring a CEO, founding engineer, QA agent, video editor, and content strategist inside Paperclip. Dotta shares practical tips on agent configuration, memory systems, skill installation, and the "Memento Man" mental model for keeping agents on track. The conversation covers everything from token spend management and agentic design patterns to the future of importable, shareable companies and the upcoming Maximizer Mode. Skills to build your agent team: https://startup-ideas-pod.link/skill-suite Timestamps: 00:00 Intro 02:32 What is Paperclip 04:21 Choosing a Startup Idea for the Demo 05:48 Setting Up your agents 07:51 Hiring Your First Agent and Creating a Plan 12:39 Agent Configuration and Persona Setup 17:08 Skills: Installing and Managing Agent Capabilities 21:02 How to Get Top-Quality Output from Agents 24:05 Token Spend Tracking and Subscription Usage 25:49 Agentic Design Patterns and QA Loops 29:05 Taste and Values: What AI Still Cannot Do 30:09 How Many Agents Run the Paperclip Project 32:32 Routines: Automating Recurring Agent Tasks 36:36 Who Is Using Paperclip Today 38:57 Shareable and Importable Companies 42:49 The Unproven Frontier: Do Agent Orgs Actually Work? 42:49 Maximizer Mode and What's Next 44:29 Did Dotta Expect It to Go This Viral? Key Points Paperclip is a bring-your-own-bot orchestrator: it works with Claude Code, Codex, OpenCode, and any model on OpenRouter, so you are not locked into a single provider. AI agents are "Memento Man" — they wake up capable but with zero memory, so you need heartbeat checklists, persona prompts, and written context to keep them effective. The biggest lever for quality output is encoding your own taste and values into agent skills and brand guides, because AI can do everything except know what you actually want. Agentic design patterns like engineer-to-QA review loops matter more than one-shotting an entire startup; structure prevents compounding errors. Paperclip tracks every token spent and every task completed, solving the problem of running dozens of agent windows with zero accountability. Importable, shareable company templates (like Gary Tan's G-Stack or a full game studio) point toward a future where you "aqua-hire" proven agent teams instead of building from scratch. The #1 tool to find startup ideas/trends - https://www.ideabrowser.com LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/ The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/ FIND ME ON SOCIAL X/Twitter: https://twitter.com/gregisenberg Instagram: https://instagram.com/gregisenberg/ LinkedIn: https://www.linkedin.com/in/gisenberg/ FIND DOTTA ON SOCIAL X/Twitter: https://x.com/dotta Paperclip: https://paperclip.ing Github: https://github.com/cryppadotta

Huh? Where'd you come from? Oh well. I guess you're with us on this one. Welcome to episode two tree tree of Brothers in Arms Podcast! Tonight we learn about the land of the big PX, personnel shopping, you know you've made it when you can buy uniform stuff, and a 24-hour planet smoothie, they make water you can actually drink, that one sounds like a bowel movement, I glow but not because I want to, I'm a submariner, has the alphabet at the end of his name, don't make it weird, I wouldn't lie to you - my dad's a preacher, Oh sure sure, new to the pokemon go thing, parked 7 hours in 3 hour parking, an irrational fear of being late, rest in peace, Kitty, poop blood and fur, an appointment for arthritis - that's dope, "hey Doc, is this normal?", You're like broke broke, I deny your request and submit one of my own, Paperclip, yeah no gentlemen, In Jamaica - Road is a Road, keto, a new appreciation for what you have, it'd be a crying shame, numb my gums, not my gumdrop buttons, "Not Gum Gum!", and a few Dad jokes that hit center mass! All this and a few healthy health updates on this week's episode of Brothers in Arms! Where you can reach us: YouTube: BrothersinArmsPodcast Instagram: Yourbrothersinarmspodcast Gmail: yourbrothersinarmspodcast@gmail.com Twitch: Twitch.tv/brothersinarmspodcast (schedule varies due to life) Website: https://brothersinarms.podbean.com

Episode 7: Understanding Maternal Ambivalence In this episode of Paperclips & Periods, hosts Dr. Emily K. Cabrera, PMHNP-BC, and Katharine "Katie" Krych, MSN, RN, tackle a topic many mothers feel but rarely say out loud: maternal ambivalence—loving your children while longing for your pre-parent self. Emily and Katie create a safe space to explore what maternal ambivalence actually means—and what it doesn't. Missing yourself, questioning decisions, or grieving your old life does not mean you regret your children or are a bad parent. These feelings reflect the profound identity shift that comes with caregiving. You weren't born "Mom"—you were born you. The hosts explain that maternal ambivalence often stems from exhaustion, sleep deprivation, and the relentless mental load—not from wanting to undo parenthood. They distinguish between normal feelings and clinical concerns like postpartum depression, providing guidance on when to seek professional help. They also validate mothers who have experienced late-term miscarriage, emphasizing that grief and hormonal shifts require support even when a baby isn't physically present. Throughout the episode, Katie and Emily challenge the cultural expectation that mothers should be superhuman and always put-together. They normalize the guilt of asking for help, the fear of judgment, and the difficulty of trusting others with your children. Using techniques from Cognitive Behavioral Therapy and Acceptance and Commitment Therapy, they encourage listeners to reframe thoughts: replace "I shouldn't feel this way" with "I love my child, and it's okay that I miss my old life." The hosts guide listeners through practical exercises: identify what you miss most, reclaim small rituals that reconnect you with yourself, and recognize when feelings cross into clinical depression. They share personal strategies—taking baths with the door locked, grounding in nature, finding moments of rest—demonstrating that self-care doesn't require grand gestures, just intentional moments. The episode closes with the signature box breathing exercise—a 16-second nervous system reset. Reflective and validating, this conversation invites mothers to release the myth of perfection and embrace the truth: you can love your children fiercely and still miss parts of yourself. You don't have to carry everything alone. Paperclips & Periods airs on Dreamvisions 7 Radio Network, a Boston-based syndicated internet radio station reaching listeners across 135 to 200+ countries through platforms including iHeartRadio, TuneIn, Stitcher, Spotify, and more. The podcast aligns with the mission of Dual Minds Integrative Psychiatry, supporting conversations that promote emotional well-being, maternal mental health, and whole-person care. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

Episode 6: The Mental Load I Didn't Consent For – Paperclips & Periods Podcast In this episode of Paperclips & Periods, hosts Dr. Emily K. Cabrera, PMHNP-BC, and Katharine "Katie" Krych, MSN, RN, dive into the invisible burden exhausting women everywhere: the mental load. This is the constant cognitive labor of remembering, planning, organizing, and managing every detail of household and family life without recognition or rest. Emily and Katie unpack what the mental load actually looks like. It's not just doing the laundry. It's remembering the laundry needs to be done, noticing when detergent is running low, adding it to the shopping list, and keeping track of who needs clean uniforms tomorrow. It's being the household manager, the default parent, the one holding everyone's schedules while your own needs fall to the bottom. The hosts explore why this burden falls disproportionately on women, even when partners "help." They discuss how helping is not the same as owning the responsibility, and why being asked to delegate tasks you never agreed to manage creates resentment and burnout. Katie and Emily validate the exhaustion of carrying invisible labor, the guilt of feeling ungrateful, and the anger of shouldering a load you never consented to carry alone. This conversation extends beyond motherhood to nurses, first responders, and healthcare workers managing professional and personal lives simultaneously. The hosts explain how the mental load compounds stress, contributes to anxiety and depression, and often goes unrecognized until someone breaks. Katie and Emily offer practical strategies: name the mental load out loud, have honest conversations with partners about shared ownership (not just shared tasks), and set boundaries. They discuss tools like shared calendars and the concept of letting go when others do things differently than you would. The hosts also address when the mental load becomes a mental health crisis, providing guidance on recognizing burnout and caregiver fatigue. Emily explains how therapy can help process resentment, rebuild boundaries, and reclaim mental space. The episode closes with the signature box breathing exercise. Honest and validating, this conversation gives permission to acknowledge the invisible work you do and challenges the expectation that you should carry it all. You deserve partnership, not help. You deserve rest, not just productivity. And you deserve to be seen. Paperclips & Periods airs on Dreamvisions 7 Radio Network, a Boston-based syndicated internet radio station reaching listeners across 135 to 200+ countries through platforms including iHeartRadio, TuneIn, Stitcher, Spotify, and more. The podcast aligns with the mission of Dual Minds Integrative Psychiatry, supporting conversations that promote emotional well-being, maternal mental health, and whole-person care. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

Episode 5: PMS, Hormones & the Emotional Realities of Womanhood  In this episode of Paperclips & Periods, hosts Dr. Emily K. Cabrera, EdD, MSN, CAGS, PMHNP‑BC, and Katharine “Katie” Krych, MSN, RN, Graduate Certificate in Nursing Education, PEL‑CSN, open up an honest, unfiltered conversation about one of the most universal yet least openly discussed aspects of women's lives: hormones—from first periods to perimenopause, and every emotional, physical, and psychological shift in between. Together, they explore how hormonal changes shape women's daily experiences, communication, relationships, and mental health across the lifespan. The discussion moves naturally from early puberty and helping young girls understand their bodies, to the complexities of PMS and unpredictable mood shifts, to the emotional impact of fertility struggles, pregnancy loss, and postpartum changes. With vulnerability and humor, they share their personal stories as mothers, clinicians, partners, and women navigating the evolution of their own cycles. Drawing from their backgrounds in psychiatric mental health, nursing, and education, Emily and Katie unpack the emotional realities behind menstruation and reproductive transitions—how cycles sync, how hormones influence sensitivity and emotional regulation, how cultural messaging shapes young girls' understanding of their bodies, and how women often carry the invisible weight of silence when navigating infertility, loss, or perimenopause. They also highlight the layered challenges nurses and caregivers face when balancing their clinical knowledge with their lived emotional experiences. This episode explores the private struggles that often accompany womanhood, including the monthly disappointment of a period when trying to conceive, the loneliness of maintaining secrecy after pregnancy loss, the fear and anxiety during high‑risk pregnancies, and the unexpected emotional reactivity that can surface during perimenopause. The hosts examine how partners cope differently, how miscommunication can deepen isolation, and why many women feel unsupported during some of the most physically and emotionally demanding moments of their lives. Grounded in lived experience, emotional honesty, and clinical insight, this episode reframes hormonal health as far more than a physical process—it is a deeply human journey that deserves openness, compassion, and community. Emily and Katie emphasize the need for generational change, encouraging listeners to speak truthfully about their experiences and to teach their children healthier ways to understand their bodies, emotions, and boundaries. As always, the hosts offer grounding takeaways, including the importance of support networks, the value of speaking openly with trusted others, and the need for emotional follow‑up during fertility challenges and pregnancy loss—areas where the healthcare system often falls short. The episode closes with a calming moment of box breathing to help listeners regulate their nervous systems and reconnect to their bodies with gentleness. Reflective, validating, and deeply real, this episode invites women to honor the full emotional landscape of their hormonal lives—and reminds every listener: you do not have to navigate these experiences in silence. Paperclips & Periods airs on the Dreamvisions 7 Radio Network, a Boston‑based syndicated internet radio station reaching listeners across 135–200+ countries through platforms including iHeartRadio, TuneIn, Stitcher, Spotify, and others. The podcast aligns with the mission of Dual Minds Integrative Psychiatry, supporting conversations that promote emotional well‑being, personal growth, and whole‑person care. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

Episode 4: High‑Achieving Women Who Are Silently Struggling In this episode of Paperclips & Periods, hosts Dr. Emily K. Cabrera, EdD, MSN, CAGS, PMHNP‑BC, and Katharine “Katie” Krych, MSN, RN, Graduate Certificate in Nursing Education, PEL‑CSN, sit down with special guest Dr. Arlicia Miller, founder and Chief Transformation Officer of the Umbrella of Artistic Expression and life transformation coach with Dual Minds Integrative Psychiatry. Together, they dive into a topic so many women live with daily but rarely name out loud: the experience of being a high‑functioning woman who is quietly, persistently struggling beneath the surface. Drawing from Dr. Miller's transformational coaching work, Dr. Cabrera's psychiatric mental health expertise, and Katie's background in nursing and education, the conversation unpacks the hidden challenges that accompany competence, ambition, caregiving roles, and emotional labor. They explore why high‑functioning women often feel obligated to “push through,” how early conditioning reinforces silence, and why vulnerability can feel risky—even among friends, colleagues, and partners. From motherhood and marriage, to career advancement, to the weight of societal expectations, this episode explores how women learn to hold everything together externally while internally battling exhaustion, depletion, and self‑doubt. The hosts also examine how gender norms, family roles, trauma histories, and cultural narratives shape women's measurements of worthiness and success. With compassion and honesty, they discuss the “struggle bus,” the fear of judgment, the stigma around asking for help, and the labels women often carry without realizing how deeply they shape identity. Grounded in lived experience, psychology, and integrative wellness, this episode reframes “high‑functioning” not as a badge of honor, but as a clue—an invitation to pause, rest, and reconnect with one's authentic self. As always, the hosts offer thoughtful takeaways and practical strategies, including the importance of small resets, the power of journaling, the need for safe relationships, and even a guided moment of box breathing to help listeners regulate their nervous systems in real time. Reflective, relatable, and deeply human, this episode encourages women to release the myth of having it all together and replace it with a more compassionate truth: you don't have to carry everything alone. Dreamvisions 7 Radio Network syndicates content widely, partnering with dozens of platforms and directories (including TuneIn, iHeartRadio, Stitcher, Spotify, and more), giving Paperclips & Periods ongoing global exposure beyond traditional podcast outlets.Paperclips & Periods aligns with the mission of Dual Minds Integrative Psychiatry, supporting whole-person care and conversations that promote emotional well-being, understanding, and growth. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

Episode 3: Setting Boundaries Without Becoming the Villain – Paperclips & Periods Podcast In this episode of Paperclips & Periods, the conversation focuses on one of the most challenging and misunderstood topics for women: boundaries. Hosted by Dr. Emily K. Cabrera, EdD, MSN, CAGS, PMHNP-BC, and Katharine “Katie” Krych, MSN, RN, Graduate Certificate in Nursing Education, PEL-CSN, this episode features Dr. Jamy Gaynor, EdD, MS, RN, NCSN, MSN(c), a neuroscience-trained school nurse whose work centers on child development, emotional regulation, and nervous system awareness. Drawing from Jamy's experience working closely with children, families, and school systems, alongside Dr. Cabrera's background in psychiatric mental health and Katie's experience in nursing and education, the conversation explores how boundaries are shaped by caregiving roles, trauma, and social conditioning — and why women are often penalized for setting them. This episode examines how boundary challenges show up across the lifespan, from childhood and adolescence to adult personal and professional relationships. Particular attention is given to how children, especially girls, internalize messages about compliance, emotional labor, and self-advocacy. Grounded in psychology, neuroscience, and lived experience, the discussion reframes boundaries not as rejection or conflict, but as essential practices for safety, clarity, and self-respect. Thoughtful, reflective, and intentionally human, this episode invites listeners to reconsider what it means to hold boundaries — and why doing so is an act of care for ourselves and future generations. Paperclips & Periods is broadcast on the Dreamvisions 7 Radio Network, a Boston-based syndicated internet radio station with a global reach. The network streams shows locally, nationally, and internationally — with listeners in well over 135 countries around the world, and in some listings even 200+ countries across platforms and syndication partners. Dreamvisions 7 Radio Network syndicates content widely, partnering with dozens of platforms and directories (including TuneIn, iHeartRadio, Stitcher, Spotify, and more), giving Paperclips & Periods ongoing global exposure beyond traditional podcast outlets.Paperclips & Periods aligns with the mission of Dual Minds Integrative Psychiatry, supporting whole-person care and conversations that promote emotional well-being, understanding, and growth. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio Paperclips & Periods Podcast  paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

Episode 2: Navigating Trauma, Emotional Blunting, and Resilience In Episode 2 of Paperclips & Periods, hosts Dr. Emily K. Cabrera, EdD, MSN, CAGS, PMHNP-BC, and Katharine “Katie” Krych, MSN, RN, PEL-CSN, dive into the real-life challenges of navigating past trauma as both moms and healthcare professionals. This episode explores emotional blunting—the way our minds sometimes shut down feelings as a coping mechanism—and how it can affect personal and professional life. Emily and Katie share insights on building resilience, emphasizing the importance of continuing forward even when the weight of past experiences feels heavy. Listeners are also introduced to box breathing, a practical tool to remain grounded during moments of stress and overwhelm. The conversation is candid, reflective, and deeply human, offering both professional perspectives and personal experiences that resonate with anyone balancing caregiving roles, motherhood, and the demands of high-pressure environments. Episode 2 reinforces the podcast's commitment to creating a space for honest, nuanced discussions around mental health, self-care, and growth. It invites listeners to explore coping strategies, acknowledge their experiences, and cultivate strength with intention and compassion. New episodes of Paperclips & Periods air every Friday at 7:00 a.m. and 7:00 p.m. Paperclips & Periods aligns with the mission of Dual Minds Integrative Psychiatry, supporting whole-person care and conversations that promote emotional well-being, understanding, and growth. Learn more: www.dualmindspsychiatry.com | Listen on Dream Visions 7 Radio

VP S6 Ep 291: Collaged Paperclips! The Junk Journal Podcast! The Paper Outpost Podcast! The Joy of Junk Journals! Free to Listen Anytime! Every Tuesday & Thursday! Topics: Junk Journals, Paper Crafting, life of a crafter, answering crafty questions! Come have a listen on Apple Podcast, Spotify, Google Podcast or go to https://anchor.fm/the-paper-outpost Also check out my Video Podcasts on M,W, F, S, S on Spotify! :) You can make your own Podcast! It's easy at Anchor: Here is how!: anch.co/outpost Grab a FUNDLE! Now available in my Etsy Shop!: 100 pieces! A mix of antique/vintage ledger pages, hand-dyed papers, old postcards, tea cards, handwritten paper, awesome vintage book pages and so much more! Wonderful to use in your junk journal creations! Free Priority Shipping in the USA! :) Limited supply! :) See a Fundle Video!:) https://youtu.be/KJnWd9RSpOQ Buy a Fundle! :) Etsy Shop: https://www.etsy.com/listing/1007331616/antique-vintage-ephemera-paper?ref=shop_home_active_6&frs=1&crt=1 VINTAGE DIGIKITS! Amazing images to download & print out at home on your printer!: Etsy Shop: https://www.etsy.com/shop/ThePaperOutpost PRINT & MAIL Option for Vintage Digikits! :) I heard your call :) No Printer? No Problem! :) I will print & mail 10 Digikits to you! Free Priority Shipping in the USA! :) 1. Select 10 names of digikits, & send me the list via Etsy message or email to pam@thepaperoutpost.com or simply say "Surprise me!" :) 2. Then buy the Print & Mail Digikit option in my Etsy shop! :) Direct Link to Buy here: https://www.etsy.com/listing/1071078687/printed-mailed-digikits-no-printer?ref=shop_home_active_1&frs=1&crt=1 That's 50 Pages total on lightweight cardstock! See All My Digikits! https://www.etsy.com/shop/ThePaperOutpost Sincerely, Pam at The Paper Outpost :)!! I am currently buried in paper and covered in glue ;) Remember that Fun Can Be Simple! Go Forth and Create with Reckless Abandon! :) MY AMAZON STORE!: My Personal Favorite Products & Tools!: Click here to see all my items in one click with pictures in my Amazon Store! https://www.amazon.com/shop/thepaperoutpost NEWSLETTER!: Free Monthly Emailed Newsletter from The Paper Outpost! Sign Up here: https://bit.ly/paperoutpostnewsletter - Free Monthly Digital Printable! - Free The Note From The Book Maker explaining what a junk journal is and how to use it! - Free Page List of Ideas for Junk Journals! - Free Checklist of Junk Journal Supplies! - Junk Journal Tips & Updates from Pam at The Paper Outpost! COME FIND ME AT :) All My Links: https://linktr.ee/thepaperoutpost ETSY Shop: https://www.thepaperoutpost.com ETSY Shop: https://www.etsy.com/shop/ThePaperOutpost YOUTUBE: https://www.youtube.com/ThePaperOutpost NEWSLETTER: https://bit.ly/paperoutpostnewsletter INSTAGRAM: https://www.instagram.com/thepaperoutpost FACEBOOK: https://www.facebook.com/ThePaperOutpost The Paper Outpost FACEBOOK GROUP: https://www.facebook.com/ThePaperOutpost/ THE PAPER OUTPOST PODCAST: The Joy of Junk Journals!: https://anchor.fm/the-paper-outpost AMAZON STORE: https://www.amazon.com/shop/thepaperoutpost PINTEREST: https://www.pinterest.com/thepaperoutpost TWITTER: https://twitter.com/thepaperoutpost MERCHANDISE STORE!: https://the-paper-outpost-2.creator-spring.com/ #thepaperoutpost #paperoutpost #thepaperoutpostpodcast #digikits #junkjournal #junkjournals #howtomakeajunkjournal #junkjournalpodcast #thejoyofjunkjournals #fundle #thejunkjournalpodcast

Jeff & Shannon celebrate Jeff's birthday (shared with Lincoln's 217th), shred Dataset 12's coded journals exposing a teenage victim's nightmare as a human incubator in Epstein's depraved breeding program, connect Bannon's Epstein ties, and call out elite protection rackets. Tune in at Rumble, YouTube, X and Red State Talk Radio! Patriots, hold on tight—this episode packs a patriotic punch from start to finish in **Season 8, Episode 029, "The Coded Journals in the Epstein Files That Hide a Teenage Girl's Horrific Story"**! @intheMatrixxx and @shadygrooove kick things off with a massive surprise birthday bash for Jeff (@intheMatrixxx), who shares the date with Abraham Lincoln's 217th birthday. Custom songs from Mikey Mariano, AARP memes, viewer Rumble rants, gift subs, and even a special "Happy birthday, Jeffrey, you're a true patriot" text straight from President Trump set the high-energy tone. Jeff reflects on his premature birth in the Land of Lincoln, breech at 5.7 ounces, and growing up with the Emancipation Proclamation on the wall—perfect backdrop for honoring the man who kept America from being torn apart by globalist bankers. Then the show dives straight into the gut-wrenching core: **Dataset 12** from the justice.gov/Epstein files, released under the Epstein Files Transparency Act. These are the raw, coded journals of a teenage victim starting at age 16—magazine cutouts, glued birthday cards for her 16th, sonograms, Sylvia Plath poems with underlined lines, and vertically readable handwritten codes documenting years of sexual slavery, repeated forced pregnancies, and babies ripped away minutes after birth, sometimes with Ghislaine Maxwell in the room ordering her to "close your eyes." She names the monsters: Leon Black (violent rapes), George Mitchell, Ted Leonsis, Steve Case, James Kimsey, John Colgan, "The Gregorys," and more from the flights and yachts of horror. The victim describes being used as a "human incubator" for a so-called superior gene pool—Epstein's eugenics obsession tied to hair color, eye color, musical talent—feeling "very Nazi-like" in her own words. This isn't speculation; it's her documented trauma, provable in court, exposing a systematic breeding program the mainstream media blackouts while obsessing over distractions. Jeff and Shannon connect the dots they've been targeted for years over: genetics, bloodlines, MKUltra-style programming, hot/cold treatment, compartmentalized minds, and the "born in" pattern from historical operations like Paperclip. They contrast this hard evidence against Pizzagate-style noise, slam the two-tiered justice system protecting elites, and spotlight networks working overtime to silence the truth—including deep ties like Steve Bannon borrowing Epstein's plane (not friendship, but partnership and cohorts), his silence on the files, and contradictions to his public MAGA persona. They also break down the Super Bowl LIX Bad Bunny halftime viewership crash (nearly 20% drops in markets like Boston and Seattle), elite disconnect, virtue-signaling rage bait, and the NFL's shame—plus a quick hat tip to Pam Bondi's congressional "Anons" gesture on Brennan indictments. After 8 years of fighting censorship, deplatforming, and attacks for exposing bloodline control and child protection, the message is clear: enough is enough. The truth is learned, never told. The constitution is your weapon. Tune in at noon-0-five Eastern LIVE to stand with Trump! MG Show: America First MAGA Podcast & Conservative Talk Show Launched in 2019 and now in Season 8, the MG Show is your go-to source for unfiltered truth on Trump policies, border security, economic nationalism, and exposing globalist psyops. Hosted by Jeffrey Pedersen (@InTheMatrixxx) and Shannon Townsend (@ShadyGrooove), it champions sovereignty, traditional values, and critiques of establishment politics. Tune in weekdays at 12pm ET / 9am PT for patriotic insights strengthening the Republic under President Trump's America First agenda. Hosts - Jeffrey Pedersen (@InTheMatrixxx): Expert in political analysis and exposing hidden agendas, with a focus on Trump's diplomatic wins and media bias. - Shannon Townsend (@ShadyGrooove): Delivers sharp insights on intelligence operations, Constitutional rights, and defenses of Trump's strategies against mainstream critiques. Where to Watch & Listen Catch live episodes or on-demand replays packed with MAGA victories like inflation drops, border awards, Trump pardons, and psyop exposures: - Live Streams: https://rumble.com/mgshow for premium America First content. - Radio: https://mgshow.link/redstate on Red State Talk Radio. - X Live: https://x.com/inthematrixxx for real-time pro-Trump discussions. - Podcasts: Search "MG Show" on PodBean, Apple Podcasts, Pandora, and Amazon Music. - YouTube: Full episodes at https://youtube.com/c/inthematrixxx and https://www.youtube.com/c/TruthForFreedom. Follow for daily pro-Trump alerts: - X: @InTheMatrixxx (https://x.com/inthematrixxx) and @ShadyGrooove (https://x.com/shadygrooove). Support the MG Show Fuel the MAGA movement against establishment lies: - Donate: https://mg.show/support or contribute at https://givesendgo.com/helpmgshow. - Merch: https://merch.mg.show for official gear. - MyPillow Special: Use code MGSHOW at https://mypillow.com/mgshow. - Crypto: https://mgshow.link/rumblewallet. All Links Everything MG Show Related: https://linktr.ee/mgshow. MG Show Anthem Get chills with the patriotic track: https://youtu.be/SyfI8_fnCAs

Step Into an Honest, Curiosity-driven Discussion Paperclips & Periods Podcast officially launches with Episode 1, an introductory conversation that sets the tone and intention for the series. Hosted by Dr. Emily K. Cabrera, EdD, MSN, CAGS, PMHNP-BC, co-founder of Dual Minds Integrative Psychiatry, and Katharine “Katie” Krych, MSN, RN, Graduate Certificate in Nursing Education, PEL-CSN, this inaugural episode focuses on why the podcast was created, who it is meant to serve, and the kinds of conversations listeners can expect going forward. Step into an honest, curiosity-driven discussion that introduces the heart of Paperclips & Periods – where mental health, mom-hood, women-hood, and real life intersect. Drawing from their combined backgrounds in nursing and education, Dr. Cabrera, PMHNP-BC and Katie, RN outline the intent behind the podcast and the values that will guide future conversations. Grounded, reflective, and intentionally human, Episode 1 invites listeners into a space designed for thoughtful dialogue rather than quick fixes. It serves as a foundation for exploring complex topics with nuance, compassion, and clarity. New episodes of Paperclips & Periods will air Fridays at 7:00 a.m. and 7:00 p.m. each week. Paperclips & Periods aligns with the mission of Dual Minds Integrative Psychiatry, supporting whole-person care and conversations that promote emotional well-being, understanding, and growth. Paperclips & Periods Podcast  paperclipsandperiods@gmail.com Dual Minds Integrative Psychiatry www.dualmindspsychiatry.com

  Multiple VS Code windows. "Agent stopping" in a robot voice. A laptop stand on the treadmill so Claude can keep working while I run. The Big Rich sitting unread by the fireplace while I check if the migration's done. Somewhere along the way, I started reorganizing my life around keeping the machine spinning. Claude Code had become my universal paperclip clicker. This is me trying to figure out the difference between real work and just feeding it tickets.   This is some field notes, a shorter, rougher than a normal epsidoe.  Episode Page Support The Show Subscribe To The Podcast Join The Newsletter

À la fin de la Seconde Guerre mondiale, l'Europe est en ruines, l'Allemagne vaincue, et le monde découvre l'ampleur des crimes du régime nazi. Pourtant, dans l'ombre des procès et des dénazifications officielles, une autre histoire commence. Une histoire secrète, pragmatique, et profondément troublante : l'opération Paperclip.Nous sommes en 1945. Les États-Unis comprennent rapidement que la victoire militaire n'est qu'une étape. Un nouveau conflit se profile déjà : la rivalité avec l'Union soviétique. Dans cette course à la puissance, un trésor attire toutes les convoitises : les scientifiques allemands. L'Allemagne nazie, malgré sa défaite, possède certains des ingénieurs et chercheurs les plus avancés du monde, notamment dans les domaines des fusées, de l'aéronautique, de la chimie et de la médecine.Washington décide alors d'agir vite. Très vite.L'opération Paperclip est lancée dans le plus grand secret. Son objectif : identifier, recruter et transférer aux États-Unis des centaines de scientifiques allemands, même lorsque leur passé est entaché d'une collaboration active avec le régime nazi.Le nom « Paperclip », trombone en anglais, vient d'une pratique administrative simple mais lourde de sens : on agrafe aux dossiers compromettants une nouvelle fiche « nettoyée », supprimant toute mention trop gênante du passé politique de certains candidats.Parmi ces recrues figure un nom devenu célèbre : Wernher von Braun. Ingénieur vedette du programme de missiles V2, armes qui ont semé la terreur à Londres et Anvers, il est récupéré avec son équipe et installé aux États-Unis. Quelques années plus tard, cet ancien scientifique du IIIᵉ Reich devient l'un des architectes du programme spatial américain et contribue directement à l'envoi des astronautes sur la Lune.Mais Paperclip ne se limite pas aux fusées. Médecins, chimistes, spécialistes en armement, chercheurs en électronique ou en sous-marins traversent eux aussi l'Atlantique. Officiellement, il s'agit de protéger ces connaissances contre une récupération soviétique. Officieusement, on ferme souvent les yeux sur des zones d'ombre : travail forcé, proximité avec la SS, expérimentations humaines.Le dilemme est immense. D'un côté, une exigence morale : juger les responsables des crimes nazis. De l'autre, une logique stratégique : ne pas laisser ces cerveaux tomber aux mains de Moscou.Entre 1945 et le début des années 1950, plus de 1 600 scientifiques allemands sont ainsi transférés vers les États-Unis grâce à Paperclip.Cette opération contribue directement à la supériorité technologique américaine pendant la Guerre froide : missiles balistiques, aviation supersonique, et bien sûr conquête spatiale.L'opération Paperclip révèle une vérité dérangeante : dans certaines circonstances, les grandes puissances sont prêtes à sacrifier la justice sur l'autel de la puissance. Une page sombre et paradoxale de l'histoire, où les anciens ennemis deviennent des alliés… au nom de l'avenir. Hébergé par Acast. Visitez acast.com/privacy pour plus d'informations.

IX En este episodio de La Llamada de la Luna exploramos una de las historias más inquietantes de la cultura contemporánea: la posible conexión entre la serie Stranger Things y supuestos proyectos militares secretos desarrollados en Estados Unidos durante el siglo XX. La investigación nos lleva hasta 1992, cuando Preston B. Nichols y Peter Moon publicaron el libro The Montauk Project: Experiments in Time, una obra que planteaba la existencia de experimentos clandestinos relacionados con el control mental, la manipulación del tiempo y la apertura de portales interdimensionales. Para comprender el origen de estas teorías, retrocedemos a 1943, al llamado Experimento Filadelfia (Project Rainbow), donde, según testimonios, la Marina estadounidense habría intentado volver invisible al destructor USS Eldridge mediante campos electromagnéticos, provocando consecuencias imposibles de explicar desde la física convencional. Décadas después, estas investigaciones habrían derivado en el Proyecto Phoenix y posteriormente, en el Proyecto Montauk, desarrollado supuestamente en la base militar de Camp Hero, en Montauk Point, Long Island. Allí se habrían realizado experimentos con radar, frecuencias electromagnéticas, control mental y supuestas capacidades psíquicas amplificadas mediante la llamada “Silla Montauk”. Entre los nombres vinculados a esta narrativa aparecen figuras como Duncan Cameron, señalado como sujeto psíquico principal; John von Neumann, presunto director científico del proyecto; Stewart Swerdlow y Christopher Loffreno, quienes afirmaron haber sido víctimas de experimentos; y Al Bielek, supuesto testigo de los acontecimientos más extremos del programa. El relato culmina el 12 de agosto de 1983, fecha en la que, según los testimonios, un experimento habría salido de control, materializando una entidad desconocida y provocando el colapso del proyecto. Pero esta historia no puede entenderse sin su contexto histórico real. Programas como MK-Ultra (1953–1973), revelado oficialmente por el Congreso estadounidense en 1975, demostraron que el gobierno sí realizó experimentos ilegales de control mental con ciudadanos sin su consentimiento. A ello se suman operaciones como Paperclip, mediante la cual más de 1.600 científicos nazis fueron trasladados a Estados Unidos tras la Segunda Guerra Mundial para trabajar en proyectos militares y científicos. La influencia del mito de Montauk llegó incluso a la cultura popular. Los hermanos Matt y Ross Duffer desarrollaron originalmente Stranger Things bajo el título Montauk, inspirándose en estas teorías, en los supuestos experimentos con niños y en la idea de portales a otras dimensiones. El episodio también aborda casos contemporáneos como la muerte del investigador británico Maxwell Bates-Spiers en 2016, cuyos mensajes previos y circunstancias alimentaron nuevas sospechas sobre proyectos secretos y redes ocultas. El Proyecto Montauk carece de pruebas verificables, pero surge en un contexto histórico donde los experimentos secretos, el control mental y las operaciones encubiertas sí existieron. La frontera entre mito y realidad se vuelve difusa: quizá la historia sea una construcción conspirativa, quizá una exageración de hechos reales, o quizá una combinación de ambos. Este episodio no pretende convencer, sino plantear la pregunta esencial: ¿Estamos ante una ficción elaborada o ante fragmentos de una verdad que nunca fue revelada? El expediente permanece abierto. Escúchanos en iVoox | Suscríbete en tu plataforma preferida HAZTE MECENAS: No dejes que La Biblioteca cierre nunca sus puertas. Suscríbete en iVoox Memberial o en tu Plataforma preferida y comparte. Gracias a los MECENAS: sin ustedes, La Llamada De La Luna no sería posible. Canal Telegram: https://t.me/LaLamadaDeLaLuna YouTube: https://www.youtube.com/channel/UCEOtdbbriLqUfBtjs_wtEHw Escucha el episodio completo en la app de iVoox, o descubre todo el catálogo de iVoox Originals

https://youtu.be/li8ZYyFr9Hc

Click here to share your favorite car, car story or any automotive trivia!What do you do when you retire from a corporate career but still want to build things? If you're Mark Freibaum, you buy a 24-foot NASCAR auto hauler and transform it into a mobile woodworking shop for kids.

Hey everyone,Fresh out the Reactor this week, we got new tunes from ENTA , Audio, Frosta, Double Medley & Subminderz, Instinkt and moreIn the Demo room we are looking at upcoming heat from Sindicate , Jocasta & Joppa , Lyness, Paperclip , Screamarts and NeurotikumCheck out the track list below and let's dive in!Stonx Music Annual 2025cygnusmusic.link/nanpoe3TRACKLIST AND MORE INFO: www.stonxmusic.co.uk/stonxcast-ep170

Basking in the glory of his first NASCAR Cup victory, rookie Dale Earnhardt finds himself in the conversation of the top talents in the garage area. Not only has he put the heat on the rest of the rookie class with his triumph, but he's put stock car racing's veterans on notice: the kid from Kannapolis is the real deal. But following up his win would be no easy task, as the next event on the Cup schedule would take him to the track deemed Too Tough To Tame, and the Lady in Black had many hard lessons to teach an overconfident freshman. After Darlington and his first trip to the Paperclip, Dale and the rest of the NASCAR contingent take on a grueling month of May, which includes the fastest race weekend to date at Talladega, a brutal 500-lapper at Dover and the longest contest in stock car racing, the World 600. Join us on this episode of Becoming Earnhardt as we unpack races 8 through 13 of the greatest NASCAR season ever, 1979. Our main character has found the spotlight, but it will be tested by not only some of the toughest events on the calendar but a toe-to-toe battle with none other than the King of NASCAR himself.FanDuel: Must be 21+ and present in select states (for Kansas, in affiliation with Kansas Star Casino) or 18+ and present in D.C. First online real money wager only. $5 first deposit required. Bonus issued as nonwithdrawable bonus bets, which expire 7 days after receipt. Restrictions apply. See terms at sportsbook.fanduel.com. Gambling Problem? Call 1-800-GAMBLER or visit FanDuel.com/RG. Call 1-888-789-7777 or visit ccpg.org/chat in Connecticut, or visit mdgamblinghelp.org in Maryland. Hope is here. Visit GamblingHelpLineMA.org or call (800) 327-5050 for 24/7 support in Massachusetts or call 1-877-8HOPE-NY or text HOPENY in New York. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.