Podcasts about Paxos

From rewriting Google's search stack in the early 2000s to reviving sparse trillion-parameter models and co-designing TPUs with frontier ML research, Jeff Dean has quietly shaped nearly every layer of the modern AI stack. As Chief AI Scientist at Google and a driving force behind Gemini, Jeff has lived through multiple scaling revolutions from CPUs and sharded indices to multimodal models that reason across text, video, and code.Jeff joins us to unpack what it really means to “own the Pareto frontier,” why distillation is the engine behind every Flash model breakthrough, how energy (in picojoules) not FLOPs is becoming the true bottleneck, what it was like leading the charge to unify all of Google's AI teams, and why the next leap won't come from bigger context windows alone, but from systems that give the illusion of attending to trillions of tokens.We discuss:* Jeff's early neural net thesis in 1990: parallel training before it was cool, why he believed scaling would win decades early, and the “bigger model, more data, better results” mantra that held for 15 years* The evolution of Google Search: sharding, moving the entire index into memory in 2001, softening query semantics pre-LLMs, and why retrieval pipelines already resemble modern LLM systems* Pareto frontier strategy: why you need both frontier “Pro” models and low-latency “Flash” models, and how distillation lets smaller models surpass prior generations* Distillation deep dive: ensembles → compression → logits as soft supervision, and why you need the biggest model to make the smallest one good* Latency as a first-class objective: why 10–50x lower latency changes UX entirely, and how future reasoning workloads will demand 10,000 tokens/sec* Energy-based thinking: picojoules per bit, why moving data costs 1000x more than a multiply, batching through the lens of energy, and speculative decoding as amortization* TPU co-design: predicting ML workloads 2–6 years out, speculative hardware features, precision reduction, sparsity, and the constant feedback loop between model architecture and silicon* Sparse models and “outrageously large” networks: trillions of parameters with 1–5% activation, and why sparsity was always the right abstraction* Unified vs. specialized models: abandoning symbolic systems, why general multimodal models tend to dominate vertical silos, and when vertical fine-tuning still makes sense* Long context and the illusion of scale: beyond needle-in-a-haystack benchmarks toward systems that narrow trillions of tokens to 117 relevant documents* Personalized AI: attending to your emails, photos, and documents (with permission), and why retrieval + reasoning will unlock deeply personal assistants* Coding agents: 50 AI interns, crisp specifications as a new core skill, and how ultra-low latency will reshape human–agent collaboration* Why ideas still matter: transformers, sparsity, RL, hardware, systems — scaling wasn't blind; the pieces had to multiply togetherShow Notes:* Gemma 3 Paper* Gemma 3* Gemini 2.5 Report* Jeff Dean's “Software Engineering Advice fromBuilding Large-Scale Distributed Systems” Presentation (with Back of the Envelope Calculations)* Latency Numbers Every Programmer Should Know by Jeff Dean* The Jeff Dean Facts* Jeff Dean Google Bio* Jeff Dean on “Important AI Trends” @Stanford AI Club* Jeff Dean & Noam Shazeer — 25 years at Google (Dwarkesh)—Jeff Dean* LinkedIn: https://www.linkedin.com/in/jeff-dean-8b212555* X: https://x.com/jeffdeanGoogle* https://google.com* https://deepmind.googleFull Video EpisodeTimestamps00:00:04 — Introduction: Alessio & Swyx welcome Jeff Dean, chief AI scientist at Google, to the Latent Space podcast00:00:30 — Owning the Pareto Frontier & balancing frontier vs low-latency models00:01:31 — Frontier models vs Flash models + role of distillation00:03:52 — History of distillation and its original motivation00:05:09 — Distillation's role in modern model scaling00:07:02 — Model hierarchy (Flash, Pro, Ultra) and distillation sources00:07:46 — Flash model economics & wide deployment00:08:10 — Latency importance for complex tasks00:09:19 — Saturation of some tasks and future frontier tasks00:11:26 — On benchmarks, public vs internal00:12:53 — Example long-context benchmarks & limitations00:15:01 — Long-context goals: attending to trillions of tokens00:16:26 — Realistic use cases beyond pure language00:18:04 — Multimodal reasoning and non-text modalities00:19:05 — Importance of vision & motion modalities00:20:11 — Video understanding example (extracting structured info)00:20:47 — Search ranking analogy for LLM retrieval00:23:08 — LLM representations vs keyword search00:24:06 — Early Google search evolution & in-memory index00:26:47 — Design principles for scalable systems00:28:55 — Real-time index updates & recrawl strategies00:30:06 — Classic “Latency numbers every programmer should know”00:32:09 — Cost of memory vs compute and energy emphasis00:34:33 — TPUs & hardware trade-offs for serving models00:35:57 — TPU design decisions & co-design with ML00:38:06 — Adapting model architecture to hardware00:39:50 — Alternatives: energy-based models, speculative decoding00:42:21 — Open research directions: complex workflows, RL00:44:56 — Non-verifiable RL domains & model evaluation00:46:13 — Transition away from symbolic systems toward unified LLMs00:47:59 — Unified models vs specialized ones00:50:38 — Knowledge vs reasoning & retrieval + reasoning00:52:24 — Vertical model specialization & modules00:55:21 — Token count considerations for vertical domains00:56:09 — Low resource languages & contextual learning00:59:22 — Origins: Dean's early neural network work01:10:07 — AI for coding & human–model interaction styles01:15:52 — Importance of crisp specification for coding agents01:19:23 — Prediction: personalized models & state retrieval01:22:36 — Token-per-second targets (10k+) and reasoning throughput01:23:20 — Episode conclusion and thanksTranscriptAlessio Fanelli [00:00:04]: Hey everyone, welcome to the Latent Space podcast. This is Alessio, founder of Kernel Labs, and I'm joined by Swyx, editor of Latent Space. Shawn Wang [00:00:11]: Hello, hello. We're here in the studio with Jeff Dean, chief AI scientist at Google. Welcome. Thanks for having me. It's a bit surreal to have you in the studio. I've watched so many of your talks, and obviously your career has been super legendary. So, I mean, congrats. I think the first thing must be said, congrats on owning the Pareto Frontier.Jeff Dean [00:00:30]: Thank you, thank you. Pareto Frontiers are good. It's good to be out there.Shawn Wang [00:00:34]: Yeah, I mean, I think it's a combination of both. You have to own the Pareto Frontier. You have to have like frontier capability, but also efficiency, and then offer that range of models that people like to use. And, you know, some part of this was started because of your hardware work. Some part of that is your model work, and I'm sure there's lots of secret sauce that you guys have worked on cumulatively. But, like, it's really impressive to see it all come together in, like, this slittily advanced.Jeff Dean [00:01:04]: Yeah, yeah. I mean, I think, as you say, it's not just one thing. It's like a whole bunch of things up and down the stack. And, you know, all of those really combine to help make UNOS able to make highly capable large models, as well as, you know, software techniques to get those large model capabilities into much smaller, lighter weight models that are, you know, much more cost effective and lower latency, but still, you know, quite capable for their size. Yeah.Alessio Fanelli [00:01:31]: How much pressure do you have on, like, having the lower bound of the Pareto Frontier, too? I think, like, the new labs are always trying to push the top performance frontier because they need to raise more money and all of that. And you guys have billions of users. And I think initially when you worked on the CPU, you were thinking about, you know, if everybody that used Google, we use the voice model for, like, three minutes a day, they were like, you need to double your CPU number. Like, what's that discussion today at Google? Like, how do you prioritize frontier versus, like, we have to do this? How do we actually need to deploy it if we build it?Jeff Dean [00:02:03]: Yeah, I mean, I think we always want to have models that are at the frontier or pushing the frontier because I think that's where you see what capabilities now exist that didn't exist at the sort of slightly less capable last year's version or last six months ago version. At the same time, you know, we know those are going to be really useful for a bunch of use cases, but they're going to be a bit slower and a bit more expensive than people might like for a bunch of other broader models. So I think what we want to do is always have kind of a highly capable sort of affordable model that enables a whole bunch of, you know, lower latency use cases. People can use them for agentic coding much more readily and then have the high-end, you know, frontier model that is really useful for, you know, deep reasoning, you know, solving really complicated math problems, those kinds of things. And it's not that. One or the other is useful. They're both useful. So I think we'd like to do both. And also, you know, through distillation, which is a key technique for making the smaller models more capable, you know, you have to have the frontier model in order to then distill it into your smaller model. So it's not like an either or choice. You sort of need that in order to actually get a highly capable, more modest size model. Yeah.Alessio Fanelli [00:03:24]: I mean, you and Jeffrey came up with the solution in 2014.Jeff Dean [00:03:28]: Don't forget, L'Oreal Vinyls as well. Yeah, yeah.Alessio Fanelli [00:03:30]: A long time ago. But like, I'm curious how you think about the cycle of these ideas, even like, you know, sparse models and, you know, how do you reevaluate them? How do you think about in the next generation of model, what is worth revisiting? Like, yeah, they're just kind of like, you know, you worked on so many ideas that end up being influential, but like in the moment, they might not feel that way necessarily. Yeah.Jeff Dean [00:03:52]: I mean, I think distillation was originally motivated because we were seeing that we had a very large image data set at the time, you know, 300 million images that we could train on. And we were seeing that if you create specialists for different subsets of those image categories, you know, this one's going to be really good at sort of mammals, and this one's going to be really good at sort of indoor room scenes or whatever, and you can cluster those categories and train on an enriched stream of data after you do pre-training on a much broader set of images. You get much better performance. If you then treat that whole set of maybe 50 models you've trained as a large ensemble, but that's not a very practical thing to serve, right? So distillation really came about from the idea of, okay, what if we want to actually serve that and train all these independent sort of expert models and then squish it into something that actually fits in a form factor that you can actually serve? And that's, you know, not that different from what we're doing today. You know, often today we're instead of having an ensemble of 50 models. We're having a much larger scale model that we then distill into a much smaller scale model.Shawn Wang [00:05:09]: Yeah. A part of me also wonders if distillation also has a story with the RL revolution. So let me maybe try to articulate what I mean by that, which is you can, RL basically spikes models in a certain part of the distribution. And then you have to sort of, well, you can spike models, but usually sometimes... It might be lossy in other areas and it's kind of like an uneven technique, but you can probably distill it back and you can, I think that the sort of general dream is to be able to advance capabilities without regressing on anything else. And I think like that, that whole capability merging without loss, I feel like it's like, you know, some part of that should be a distillation process, but I can't quite articulate it. I haven't seen much papers about it.Jeff Dean [00:06:01]: Yeah, I mean, I tend to think of one of the key advantages of distillation is that you can have a much smaller model and you can have a very large, you know, training data set and you can get utility out of making many passes over that data set because you're now getting the logits from the much larger model in order to sort of coax the right behavior out of the smaller model that you wouldn't otherwise get with just the hard labels. And so, you know, I think that's what we've observed. Is you can get, you know, very close to your largest model performance with distillation approaches. And that seems to be, you know, a nice sweet spot for a lot of people because it enables us to kind of, for multiple Gemini generations now, we've been able to make the sort of flash version of the next generation as good or even substantially better than the previous generations pro. And I think we're going to keep trying to do that because that seems like a good trend to follow.Shawn Wang [00:07:02]: So, Dara asked, so it was the original map was Flash Pro and Ultra. Are you just sitting on Ultra and distilling from that? Is that like the mother load?Jeff Dean [00:07:12]: I mean, we have a lot of different kinds of models. Some are internal ones that are not necessarily meant to be released or served. Some are, you know, our pro scale model and we can distill from that as well into our Flash scale model. So I think, you know, it's an important set of capabilities to have and also inference time scaling. It can also be a useful thing to improve the capabilities of the model.Shawn Wang [00:07:35]: And yeah, yeah, cool. Yeah. And obviously, I think the economy of Flash is what led to the total dominance. I think the latest number is like 50 trillion tokens. I don't know. I mean, obviously, it's changing every day.Jeff Dean [00:07:46]: Yeah, yeah. But, you know, by market share, hopefully up.Shawn Wang [00:07:50]: No, I mean, there's no I mean, there's just the economics wise, like because Flash is so economical, like you can use it for everything. Like it's in Gmail now. It's in YouTube. Like it's yeah. It's in everything.Jeff Dean [00:08:02]: We're using it more in our search products of various AI mode reviews.Shawn Wang [00:08:05]: Oh, my God. Flash past the AI mode. Oh, my God. Yeah, that's yeah, I didn't even think about that.Jeff Dean [00:08:10]: I mean, I think one of the things that is quite nice about the Flash model is not only is it more affordable, it's also a lower latency. And I think latency is actually a pretty important characteristic for these models because we're going to want models to do much more complicated things that are going to involve, you know, generating many more tokens from when you ask the model to do so. So, you know, if you're going to ask the model to do something until it actually finishes what you ask it to do, because you're going to ask now, not just write me a for loop, but like write me a whole software package to do X or Y or Z. And so having low latency systems that can do that seems really important. And Flash is one direction, one way of doing that. You know, obviously our hardware platforms enable a bunch of interesting aspects of our, you know, serving stack as well, like TPUs, the interconnect between. Chips on the TPUs is actually quite, quite high performance and quite amenable to, for example, long context kind of attention operations, you know, having sparse models with lots of experts. These kinds of things really, really matter a lot in terms of how do you make them servable at scale.Alessio Fanelli [00:09:19]: Yeah. Does it feel like there's some breaking point for like the proto Flash distillation, kind of like one generation delayed? I almost think about almost like the capability as a. In certain tasks, like the pro model today is a saturated, some sort of task. So next generation, that same task will be saturated at the Flash price point. And I think for most of the things that people use models for at some point, the Flash model in two generation will be able to do basically everything. And how do you make it economical to like keep pushing the pro frontier when a lot of the population will be okay with the Flash model? I'm curious how you think about that.Jeff Dean [00:09:59]: I mean, I think that's true. If your distribution of what people are asking people, the models to do is stationary, right? But I think what often happens is as the models become more capable, people ask them to do more, right? So, I mean, I think this happens in my own usage. Like I used to try our models a year ago for some sort of coding task, and it was okay at some simpler things, but wouldn't do work very well for more complicated things. And since then, we've improved dramatically on the more complicated coding tasks. And now I'll ask it to do much more complicated things. And I think that's true, not just of coding, but of, you know, now, you know, can you analyze all the, you know, renewable energy deployments in the world and give me a report on solar panel deployment or whatever. That's a very complicated, you know, more complicated task than people would have asked a year ago. And so you are going to want more capable models to push the frontier in the absence of what people ask the models to do. And that also then gives us. Insight into, okay, where does the, where do things break down? How can we improve the model in these, these particular areas, uh, in order to sort of, um, make the next generation even better.Alessio Fanelli [00:11:11]: Yeah. Are there any benchmarks or like test sets they use internally? Because it's almost like the same benchmarks get reported every time. And it's like, all right, it's like 99 instead of 97. Like, how do you have to keep pushing the team internally to it? Or like, this is what we're building towards. Yeah.Jeff Dean [00:11:26]: I mean, I think. Benchmarks, particularly external ones that are publicly available. Have their utility, but they often kind of have a lifespan of utility where they're introduced and maybe they're quite hard for current models. You know, I, I like to think of the best kinds of benchmarks are ones where the initial scores are like 10 to 20 or 30%, maybe, but not higher. And then you can sort of work on improving that capability for, uh, whatever it is, the benchmark is trying to assess and get it up to like 80, 90%, whatever. I, I think once it hits kind of 95% or something, you get very diminishing returns from really focusing on that benchmark, cuz it's sort of, it's either the case that you've now achieved that capability, or there's also the issue of leakage in public data or very related kind of data being, being in your training data. Um, so we have a bunch of held out internal benchmarks that we really look at where we know that wasn't represented in the training data at all. There are capabilities that we want the model to have. Um, yeah. Yeah. Um, that it doesn't have now, and then we can work on, you know, assessing, you know, how do we make the model better at these kinds of things? Is it, we need different kind of data to train on that's more specialized for this particular kind of task. Do we need, um, you know, a bunch of, uh, you know, architectural improvements or some sort of, uh, model capability improvements, you know, what would help make that better?Shawn Wang [00:12:53]: Is there, is there such an example that you, uh, a benchmark inspired in architectural improvement? Like, uh, I'm just kind of. Jumping on that because you just.Jeff Dean [00:13:02]: Uh, I mean, I think some of the long context capability of the, of the Gemini models that came, I guess, first in 1.5 really were about looking at, okay, we want to have, um, you know,Shawn Wang [00:13:15]: immediately everyone jumped to like completely green charts of like, everyone had, I was like, how did everyone crack this at the same time? Right. Yeah. Yeah.Jeff Dean [00:13:23]: I mean, I think, um, and once you're set, I mean, as you say that needed single needle and a half. Hey, stack benchmark is really saturated for at least context links up to 1, 2 and K or something. Don't actually have, you know, much larger than 1, 2 and 8 K these days or two or something. We're trying to push the frontier of 1 million or 2 million context, which is good because I think there are a lot of use cases where. Yeah. You know, putting a thousand pages of text or putting, you know, multiple hour long videos and the context and then actually being able to make use of that as useful. Try to, to explore the über graduation are fairly large. But the single needle in a haystack benchmark is sort of saturated. So you really want more complicated, sort of multi-needle or more realistic, take all this content and produce this kind of answer from a long context that sort of better assesses what it is people really want to do with long context. Which is not just, you know, can you tell me the product number for this particular thing?Shawn Wang [00:14:31]: Yeah, it's retrieval. It's retrieval within machine learning. It's interesting because I think the more meta level I'm trying to operate at here is you have a benchmark. You're like, okay, I see the architectural thing I need to do in order to go fix that. But should you do it? Because sometimes that's an inductive bias, basically. It's what Jason Wei, who used to work at Google, would say. Exactly the kind of thing. Yeah, you're going to win. Short term. Longer term, I don't know if that's going to scale. You might have to undo that.Jeff Dean [00:15:01]: I mean, I like to sort of not focus on exactly what solution we're going to derive, but what capability would you want? And I think we're very convinced that, you know, long context is useful, but it's way too short today. Right? Like, I think what you would really want is, can I attend to the internet while I answer my question? Right? But that's not going to happen. I think that's going to be solved by purely scaling the existing solutions, which are quadratic. So a million tokens kind of pushes what you can do. You're not going to do that to a trillion tokens, let alone, you know, a billion tokens, let alone a trillion. But I think if you could give the illusion that you can attend to trillions of tokens, that would be amazing. You'd find all kinds of uses for that. You would have attend to the internet. You could attend to the pixels of YouTube and the sort of deeper representations that we can find. You could attend to the form for a single video, but across many videos, you know, on a personal Gemini level, you could attend to all of your personal state with your permission. So like your emails, your photos, your docs, your plane tickets you have. I think that would be really, really useful. And the question is, how do you get algorithmic improvements and system level improvements that get you to something where you actually can attend to trillions of tokens? Right. In a meaningful way. Yeah.Shawn Wang [00:16:26]: But by the way, I think I did some math and it's like, if you spoke all day, every day for eight hours a day, you only generate a maximum of like a hundred K tokens, which like very comfortably fits.Jeff Dean [00:16:38]: Right. But if you then say, okay, I want to be able to understand everything people are putting on videos.Shawn Wang [00:16:46]: Well, also, I think that the classic example is you start going beyond language into like proteins and whatever else is extremely information dense. Yeah. Yeah.Jeff Dean [00:16:55]: I mean, I think one of the things about Gemini's multimodal aspects is we've always wanted it to be multimodal from the start. And so, you know, that sometimes to people means text and images and video sort of human-like and audio, audio, human-like modalities. But I think it's also really useful to have Gemini know about non-human modalities. Yeah. Like LIDAR sensor data from. Yes. Say, Waymo vehicles or. Like robots or, you know, various kinds of health modalities, x-rays and MRIs and imaging and genomics information. And I think there's probably hundreds of modalities of data where you'd like the model to be able to at least be exposed to the fact that this is an interesting modality and has certain meaning in the world. Where even if you haven't trained on all the LIDAR data or MRI data, you could have, because maybe that's not, you know, it doesn't make sense in terms of trade-offs of. You know, what you include in your main pre-training data mix, at least including a little bit of it is actually quite useful. Yeah. Because it sort of tempts the model that this is a thing.Shawn Wang [00:18:04]: Yeah. Do you believe, I mean, since we're on this topic and something I just get to ask you all the questions I always wanted to ask, which is fantastic. Like, are there some king modalities, like modalities that supersede all the other modalities? So a simple example was Vision can, on a pixel level, encode text. And DeepSeq had this DeepSeq CR paper that did that. Vision. And Vision has also been shown to maybe incorporate audio because you can do audio spectrograms and that's, that's also like a Vision capable thing. Like, so, so maybe Vision is just the king modality and like. Yeah.Jeff Dean [00:18:36]: I mean, Vision and Motion are quite important things, right? Motion. Well, like video as opposed to static images, because I mean, there's a reason evolution has evolved eyes like 23 independent ways, because it's such a useful capability for sensing the world around you, which is really what we want these models to be. So I think the only thing that we can be able to do is interpret the things we're seeing or the things we're paying attention to and then help us in using that information to do things. Yeah.Shawn Wang [00:19:05]: I think motion, you know, I still want to shout out, I think Gemini, still the only native video understanding model that's out there. So I use it for YouTube all the time. Nice.Jeff Dean [00:19:15]: Yeah. Yeah. I mean, it's actually, I think people kind of are not necessarily aware of what the Gemini models can actually do. Yeah. Like I have an example I've used in one of my talks. It had like, it was like a YouTube highlight video of 18 memorable sports moments across the last 20 years or something. So it has like Michael Jordan hitting some jump shot at the end of the finals and, you know, some soccer goals and things like that. And you can literally just give it the video and say, can you please make me a table of what all these different events are? What when the date is when they happened? And a short description. And so you get like now an 18 row table of that information extracted from the video, which is, you know, not something most people think of as like a turn video into sequel like table.Alessio Fanelli [00:20:11]: Has there been any discussion inside of Google of like, you mentioned tending to the whole internet, right? Google, it's almost built because a human cannot tend to the whole internet and you need some sort of ranking to find what you need. Yep. That ranking is like much different for an LLM because you can expect a person to look at maybe the first five, six links in a Google search versus for an LLM. Should you expect to have 20 links that are highly relevant? Like how do you internally figure out, you know, how do we build the AI mode that is like maybe like much broader search and span versus like the more human one? Yeah.Jeff Dean [00:20:47]: I mean, I think even pre-language model based work, you know, our ranking systems would be built to start. I mean, I think even pre-language model based work, you know, our ranking systems would be built to start. With a giant number of web pages in our index, many of them are not relevant. So you identify a subset of them that are relevant with very lightweight kinds of methods. You know, you're down to like 30,000 documents or something. And then you gradually refine that to apply more and more sophisticated algorithms and more and more sophisticated sort of signals of various kinds in order to get down to ultimately what you show, which is, you know, the final 10 results or, you know, 10 results plus. Other kinds of information. And I think an LLM based system is not going to be that dissimilar, right? You're going to attend to trillions of tokens, but you're going to want to identify, you know, what are the 30,000 ish documents that are with the, you know, maybe 30 million interesting tokens. And then how do you go from that into what are the 117 documents I really should be paying attention to in order to carry out the tasks that the user has asked? And I think, you know, you can imagine systems where you have, you know, a lot of highly parallel processing to identify those initial 30,000 candidates, maybe with very lightweight kinds of models. Then you have some system that sort of helps you narrow down from 30,000 to the 117 with maybe a little bit more sophisticated model or set of models. And then maybe the final model is the thing that looks. So the 117 things that might be your most capable model. So I think it has to, it's going to be some system like that, that is really enables you to give the illusion of attending to trillions of tokens. Sort of the way Google search gives you, you know, not the illusion, but you are searching the internet, but you're finding, you know, a very small subset of things that are, that are relevant.Shawn Wang [00:22:47]: Yeah. I often tell a lot of people that are not steeped in like Google search history that, well, you know, like Bert was. Like he was like basically immediately inside of Google search and that improves results a lot, right? Like I don't, I don't have any numbers off the top of my head, but like, I'm sure you guys, that's obviously the most important numbers to Google. Yeah.Jeff Dean [00:23:08]: I mean, I think going to an LLM based representation of text and words and so on enables you to get out of the explicit hard notion of, of particular words having to be on the page, but really getting at the notion of this topic of this page or this page. Paragraph is highly relevant to this query. Yeah.Shawn Wang [00:23:28]: I don't think people understand how much LLMs have taken over all these very high traffic system, very high traffic. Yeah. Like it's Google, it's YouTube. YouTube has this like semantics ID thing where it's just like every token or every item in the vocab is a YouTube video or something that predicts the video using a code book, which is absurd to me for YouTube size.Jeff Dean [00:23:50]: And then most recently GROK also for, for XAI, which is like, yeah. I mean, I'll call out even before LLMs were used extensively in search, we put a lot of emphasis on softening the notion of what the user actually entered into the query.Shawn Wang [00:24:06]: So do you have like a history of like, what's the progression? Oh yeah.Jeff Dean [00:24:09]: I mean, I actually gave a talk in, uh, I guess, uh, web search and data mining conference in 2009, uh, where we never actually published any papers about the origins of Google search, uh, sort of, but we went through sort of four or five or six. generations, four or five or six generations of, uh, redesigning of the search and retrieval system, uh, from about 1999 through 2004 or five. And that talk is really about that evolution. And one of the things that really happened in 2001 was we were sort of working to scale the system in multiple dimensions. So one is we wanted to make our index bigger, so we could retrieve from a larger index, which always helps your quality in general. Uh, because if you don't have the page in your index, you're going to not do well. Um, and then we also needed to scale our capacity because we were, our traffic was growing quite extensively. Um, and so we had, you know, a sharded system where you have more and more shards as the index grows, you have like 30 shards. And then if you want to double the index size, you make 60 shards so that you can bound the latency by which you respond for any particular user query. Um, and then as traffic grows, you add, you add more and more replicas of each of those. And so we eventually did the math that realized that in a data center where we had say 60 shards and, um, you know, 20 copies of each shard, we now had 1200 machines, uh, with disks. And we did the math and we're like, Hey, one copy of that index would actually fit in memory across 1200 machines. So in 2001, we introduced, uh, we put our entire index in memory and what that enabled from a quality perspective was amazing. Um, and so we had more and more replicas of each of those. Before you had to be really careful about, you know, how many different terms you looked at for a query, because every one of them would involve a disk seek on every one of the 60 shards. And so you, as you make your index bigger, that becomes even more inefficient. But once you have the whole index in memory, it's totally fine to have 50 terms you throw into the query from the user's original three or four word query, because now you can add synonyms like restaurant and restaurants and cafe and, uh, you know, things like that. Uh, bistro and all these things. And you can suddenly start, uh, sort of really, uh, getting at the meaning of the word as opposed to the exact semantic form the user typed in. And that was, you know, 2001, very much pre LLM, but really it was about softening the, the strict definition of what the user typed in order to get at the meaning.Alessio Fanelli [00:26:47]: What are like principles that you use to like design the systems, especially when you have, I mean, in 2001, the internet is like. Doubling, tripling every year in size is not like, uh, you know, and I think today you kind of see that with LLMs too, where like every year the jumps in size and like capabilities are just so big. Are there just any, you know, principles that you use to like, think about this? Yeah.Jeff Dean [00:27:08]: I mean, I think, uh, you know, first, whenever you're designing a system, you want to understand what are the sort of design parameters that are going to be most important in designing that, you know? So, you know, how many queries per second do you need to handle? How big is the internet? How big is the index you need to handle? How much data do you need to keep for every document in the index? How are you going to look at it when you retrieve things? Um, what happens if traffic were to double or triple, you know, will that system work well? And I think a good design principle is you're going to want to design a system so that the most important characteristics could scale by like factors of five or 10, but probably not beyond that because often what happens is if you design a system for X. And something suddenly becomes a hundred X, that would enable a very different point in the design space that would not make sense at X. But all of a sudden at a hundred X makes total sense. So like going from a disk space index to a in memory index makes a lot of sense once you have enough traffic, because now you have enough replicas of the sort of state on disk that those machines now actually can hold, uh, you know, a full copy of the, uh, index and memory. Yeah. And that all of a sudden enabled. A completely different design that wouldn't have been practical before. Yeah. Um, so I'm, I'm a big fan of thinking through designs in your head, just kind of playing with the design space a little before you actually do a lot of writing of code. But, you know, as you said, in the early days of Google, we were growing the index, uh, quite extensively. We were growing the update rate of the index. So the update rate actually is the parameter that changed the most. Surprising. So it used to be once a month.Shawn Wang [00:28:55]: Yeah.Jeff Dean [00:28:56]: And then we went to a system that could update any particular page in like sub one minute. Okay.Shawn Wang [00:29:02]: Yeah. Because this is a competitive advantage, right?Jeff Dean [00:29:04]: Because all of a sudden news related queries, you know, if you're, if you've got last month's news index, it's not actually that useful for.Shawn Wang [00:29:11]: News is a special beast. Was there any, like you could have split it onto a separate system.Jeff Dean [00:29:15]: Well, we did. We launched a Google news product, but you also want news related queries that people type into the main index to also be sort of updated.Shawn Wang [00:29:23]: So, yeah, it's interesting. And then you have to like classify whether the page is, you have to decide which pages should be updated and what frequency. Oh yeah.Jeff Dean [00:29:30]: There's a whole like, uh, system behind the scenes that's trying to decide update rates and importance of the pages. So even if the update rate seems low, you might still want to recrawl important pages quite often because, uh, the likelihood they change might be low, but the value of having updated is high.Shawn Wang [00:29:50]: Yeah, yeah, yeah, yeah. Uh, well, you know, yeah. This, uh, you know, mention of latency and, and saving things to this reminds me of one of your classics, which I have to bring up, which is latency numbers. Every programmer should know, uh, was there a, was it just a, just a general story behind that? Did you like just write it down?Jeff Dean [00:30:06]: I mean, this has like sort of eight or 10 different kinds of metrics that are like, how long does a cache mistake? How long does branch mispredict take? How long does a reference domain memory take? How long does it take to send, you know, a packet from the U S to the Netherlands or something? Um,Shawn Wang [00:30:21]: why Netherlands, by the way, or is it, is that because of Chrome?Jeff Dean [00:30:25]: Uh, we had a data center in the Netherlands, um, so, I mean, I think this gets to the point of being able to do the back of the envelope calculations. So these are sort of the raw ingredients of those, and you can use them to say, okay, well, if I need to design a system to do image search and thumb nailing or something of the result page, you know, how, what I do that I could pre-compute the image thumbnails. I could like. Try to thumbnail them on the fly from the larger images. What would that do? How much dis bandwidth than I need? How many des seeks would I do? Um, and you can sort of actually do thought experiments in, you know, 30 seconds or a minute with the sort of, uh, basic, uh, basic numbers at your fingertips. Uh, and then as you sort of build software using higher level libraries, you kind of want to develop the same intuitions for how long does it take to, you know, look up something in this particular kind of.Shawn Wang [00:31:21]: I'll see you next time.Shawn Wang [00:31:51]: Which is a simple byte conversion. That's nothing interesting. I wonder if you have any, if you were to update your...Jeff Dean [00:31:58]: I mean, I think it's really good to think about calculations you're doing in a model, either for training or inference.Jeff Dean [00:32:09]: Often a good way to view that is how much state will you need to bring in from memory, either like on-chip SRAM or HBM from the accelerator. Attached memory or DRAM or over the network. And then how expensive is that data motion relative to the cost of, say, an actual multiply in the matrix multiply unit? And that cost is actually really, really low, right? Because it's order, depending on your precision, I think it's like sub one picodule.Shawn Wang [00:32:50]: Oh, okay. You measure it by energy. Yeah. Yeah.Jeff Dean [00:32:52]: Yeah. I mean, it's all going to be about energy and how do you make the most energy efficient system. And then moving data from the SRAM on the other side of the chip, not even off the off chip, but on the other side of the same chip can be, you know, a thousand picodules. Oh, yeah. And so all of a sudden, this is why your accelerators require batching. Because if you move, like, say, the parameter of a model from SRAM on the, on the chip into the multiplier unit, that's going to cost you a thousand picodules. So you better make use of that, that thing that you moved many, many times with. So that's where the batch dimension comes in. Because all of a sudden, you know, if you have a batch of 256 or something, that's not so bad. But if you have a batch of one, that's really not good.Shawn Wang [00:33:40]: Yeah. Yeah. Right.Jeff Dean [00:33:41]: Because then you paid a thousand picodules in order to do your one picodule multiply.Shawn Wang [00:33:46]: I have never heard an energy-based analysis of batching.Jeff Dean [00:33:50]: Yeah. I mean, that's why people batch. Yeah. Ideally, you'd like to use batch size one because the latency would be great.Shawn Wang [00:33:56]: The best latency.Jeff Dean [00:33:56]: But the energy cost and the compute cost inefficiency that you get is quite large. So, yeah.Shawn Wang [00:34:04]: Is there a similar trick like, like, like you did with, you know, putting everything in memory? Like, you know, I think obviously NVIDIA has caused a lot of waves with betting very hard on SRAM with Grok. I wonder if, like, that's something that you already saw with, with the TPUs, right? Like that, that you had to. Uh, to serve at your scale, uh, you probably sort of saw that coming. Like what, what, what hardware, uh, innovations or insights were formed because of what you're seeing there?Jeff Dean [00:34:33]: Yeah. I mean, I think, you know, TPUs have this nice, uh, sort of regular structure of 2D or 3D meshes with a bunch of chips connected. Yeah. And each one of those has HBM attached. Um, I think for serving some kinds of models, uh, you know, you, you pay a lot higher cost. Uh, and time latency, um, bringing things in from HBM than you do bringing them in from, uh, SRAM on the chip. So if you have a small enough model, you can actually do model parallelism, spread it out over lots of chips and you actually get quite good throughput improvements and latency improvements from doing that. And so you're now sort of striping your smallish scale model over say 16 or 64 chips. Uh, but as if you do that and it all fits in. In SRAM, uh, that can be a big win. So yeah, that's not a surprise, but it is a good technique.Alessio Fanelli [00:35:27]: Yeah. What about the TPU design? Like how much do you decide where the improvements have to go? So like, this is like a good example of like, is there a way to bring the thousand picojoules down to 50? Like, is it worth designing a new chip to do that? The extreme is like when people say, oh, you should burn the model on the ASIC and that's kind of like the most extreme thing. How much of it? Is it worth doing an hardware when things change so quickly? Like what was the internal discussion? Yeah.Jeff Dean [00:35:57]: I mean, we, we have a lot of interaction between say the TPU chip design architecture team and the sort of higher level modeling, uh, experts, because you really want to take advantage of being able to co-design what should future TPUs look like based on where we think the sort of ML research puck is going, uh, in some sense, because, uh, you know, as a hardware designer for ML and in particular, you're trying to design a chip starting today and that design might take two years before it even lands in a data center. And then it has to sort of be a reasonable lifetime of the chip to take you three, four or five years. So you're trying to predict two to six years out where, what ML computations will people want to run two to six years out in a very fast changing field. And so having people with interest. Interesting ML research ideas of things we think will start to work in that timeframe or will be more important in that timeframe, uh, really enables us to then get, you know, interesting hardware features put into, you know, TPU N plus two, where TPU N is what we have today.Shawn Wang [00:37:10]: Oh, the cycle time is plus two.Jeff Dean [00:37:12]: Roughly. Wow. Because, uh, I mean, sometimes you can squeeze some changes into N plus one, but, you know, bigger changes are going to require the chip. Yeah. Design be earlier in its lifetime design process. Um, so whenever we can do that, it's generally good. And sometimes you can put in speculative features that maybe won't cost you much chip area, but if it works out, it would make something, you know, 10 times as fast. And if it doesn't work out, well, you burned a little bit of tiny amount of your chip area on that thing, but it's not that big a deal. Uh, sometimes it's a very big change and we want to be pretty sure this is going to work out. So we'll do like lots of carefulness. Uh, ML experimentation to show us, uh, this is actually the, the way we want to go. Yeah.Alessio Fanelli [00:37:58]: Is there a reverse of like, we already committed to this chip design so we can not take the model architecture that way because it doesn't quite fit?Jeff Dean [00:38:06]: Yeah. I mean, you, you definitely have things where you're going to adapt what the model architecture looks like so that they're efficient on the chips that you're going to have for both training and inference of that, of that, uh, generation of model. So I think it kind of goes both ways. Um, you know, sometimes you can take advantage of, you know, lower precision things that are coming in a future generation. So you can, might train it at that lower precision, even if the current generation doesn't quite do that. Mm.Shawn Wang [00:38:40]: Yeah. How low can we go in precision?Jeff Dean [00:38:43]: Because people are saying like ternary is like, uh, yeah, I mean, I'm a big fan of very low precision because I think that gets, that saves you a tremendous amount of time. Right. Because it's picojoules per bit that you're transferring and reducing the number of bits is a really good way to, to reduce that. Um, you know, I think people have gotten a lot of luck, uh, mileage out of having very low bit precision things, but then having scaling factors that apply to a whole bunch of, uh, those, those weights. Scaling. How does it, how does it, okay.Shawn Wang [00:39:15]: Interesting. You, so low, low precision, but scaled up weights. Yeah. Huh. Yeah. Never considered that. Yeah. Interesting. Uh, w w while we're on this topic, you know, I think there's a lot of, um, uh, this, the concept of precision at all is weird when we're sampling, you know, uh, we just, at the end of this, we're going to have all these like chips that I'll do like very good math. And then we're just going to throw a random number generator at the start. So, I mean, there's a movement towards, uh, energy based, uh, models and processors. I'm just curious if you've, obviously you've thought about it, but like, what's your commentary?Jeff Dean [00:39:50]: Yeah. I mean, I think. There's a bunch of interesting trends though. Energy based models is one, you know, diffusion based models, which don't sort of sequentially decode tokens is another, um, you know, speculative decoding is a way that you can get sort of an equivalent, very small.Shawn Wang [00:40:06]: Draft.Jeff Dean [00:40:07]: Batch factor, uh, for like you predict eight tokens out and that enables you to sort of increase the effective batch size of what you're doing by a factor of eight, even, and then you maybe accept five or six of those tokens. So you get. A five, a five X improvement in the amortization of moving weights, uh, into the multipliers to do the prediction for the, the tokens. So these are all really good techniques and I think it's really good to look at them from the lens of, uh, energy, real energy, not energy based models, um, and, and also latency and throughput, right? If you look at things from that lens, that sort of guides you to. Two solutions that are gonna be, uh, you know, better from, uh, you know, being able to serve larger models or, you know, equivalent size models more cheaply and with lower latency.Shawn Wang [00:41:03]: Yeah. Well, I think, I think I, um, it's appealing intellectually, uh, haven't seen it like really hit the mainstream, but, um, I do think that, uh, there's some poetry in the sense that, uh, you know, we don't have to do, uh, a lot of shenanigans if like we fundamentally. Design it into the hardware. Yeah, yeah.Jeff Dean [00:41:23]: I mean, I think there's still a, there's also sort of the more exotic things like analog based, uh, uh, computing substrates as opposed to digital ones. Uh, I'm, you know, I think those are super interesting cause they can be potentially low power. Uh, but I think you often end up wanting to interface that with digital systems and you end up losing a lot of the power advantages in the digital to analog and analog to digital conversions. You end up doing, uh, at the sort of boundaries. And periphery of that system. Um, I still think there's a tremendous distance we can go from where we are today in terms of energy efficiency with sort of, uh, much better and specialized hardware for the models we care about.Shawn Wang [00:42:05]: Yeah.Alessio Fanelli [00:42:06]: Um, any other interesting research ideas that you've seen, or like maybe things that you cannot pursue a Google that you would be interested in seeing researchers take a step at, I guess you have a lot of researchers. Yeah, I guess you have enough, but our, our research.Jeff Dean [00:42:21]: Our research portfolio is pretty broad. I would say, um, I mean, I think, uh, in terms of research directions, there's a whole bunch of, uh, you know, open problems and how do you make these models reliable and able to do much longer, kind of, uh, more complex tasks that have lots of subtasks. How do you orchestrate, you know, maybe one model that's using other models as tools in order to sort of build, uh, things that can accomplish, uh, you know, much more. Yeah. Significant pieces of work, uh, collectively, then you would ask a single model to do. Um, so that's super interesting. How do you get more verifiable, uh, you know, how do you get RL to work for non-verifiable domains? I think it's a pretty interesting open problem because I think that would broaden out the capabilities of the models, the improvements that you're seeing in both math and coding. Uh, if we could apply those to other less verifiable domains, because we've come up with RL techniques that actually enable us to do that. Uh, effectively, that would, that would really make the models improve quite a lot. I think.Alessio Fanelli [00:43:26]: I'm curious, like when we had Noam Brown on the podcast, he said, um, they already proved you can do it with deep research. Um, you kind of have it with AI mode in a way it's not verifiable. I'm curious if there's any thread that you think is interesting there. Like what is it? Both are like information retrieval of JSON. So I wonder if it's like the retrieval is like the verifiable part. That you can score or what are like, yeah, yeah. How, how would you model that, that problem?Jeff Dean [00:43:55]: Yeah. I mean, I think there are ways of having other models that can evaluate the results of what a first model did, maybe even retrieving. Can you have another model that says, is this things, are these things you retrieved relevant? Or can you rate these 2000 things you retrieved to assess which ones are the 50 most relevant or something? Um, I think those kinds of techniques are actually quite effective. Sometimes I can even be the same model, just prompted differently to be a, you know, a critic as opposed to a, uh, actual retrieval system. Yeah.Shawn Wang [00:44:28]: Um, I do think like there, there is that, that weird cliff where like, it feels like we've done the easy stuff and then now it's, but it always feels like that every year. It's like, oh, like we know, we know, and the next part is super hard and nobody's figured it out. And, uh, exactly with this RLVR thing where like everyone's talking about, well, okay, how do we. the next stage of the non-verifiable stuff. And everyone's like, I don't know, you know, Ellen judge.Jeff Dean [00:44:56]: I mean, I feel like the nice thing about this field is there's lots and lots of smart people thinking about creative solutions to some of the problems that we all see. Uh, because I think everyone sort of sees that the models, you know, are great at some things and they fall down around the edges of those things and, and are not as capable as we'd like in those areas. And then coming up with good techniques and trying those. And seeing which ones actually make a difference is sort of what the whole research aspect of this field is, is pushing forward. And I think that's why it's super interesting. You know, if you think about two years ago, we were struggling with GSM, eight K problems, right? Like, you know, Fred has two rabbits. He gets three more rabbits. How many rabbits does he have? That's a pretty far cry from the kinds of mathematics that the models can, and now you're doing IMO and Erdos problems in pure language. Yeah. Yeah. Pure language. So that is a really, really amazing jump in capabilities in, you know, in a year and a half or something. And I think, um, for other areas, it'd be great if we could make that kind of leap. Uh, and you know, we don't exactly see how to do it for some, some areas, but we do see it for some other areas and we're going to work hard on making that better. Yeah.Shawn Wang [00:46:13]: Yeah.Alessio Fanelli [00:46:14]: Like YouTube thumbnail generation. That would be very helpful. We need that. That would be AGI. We need that.Shawn Wang [00:46:20]: That would be. As far as content creators go.Jeff Dean [00:46:22]: I guess I'm not a YouTube creator, so I don't care that much about that problem, but I guess, uh, many people do.Shawn Wang [00:46:27]: It does. Yeah. It doesn't, it doesn't matter. People do judge books by their covers as it turns out. Um, uh, just to draw a bit on the IMO goal. Um, I'm still not over the fact that a year ago we had alpha proof and alpha geometry and all those things. And then this year we were like, screw that we'll just chuck it into Gemini. Yeah. What's your reflection? Like, I think this, this question about. Like the merger of like symbolic systems and like, and, and LMS, uh, was a very much core belief. And then somewhere along the line, people would just said, Nope, we'll just all do it in the LLM.Jeff Dean [00:47:02]: Yeah. I mean, I think it makes a lot of sense to me because, you know, humans manipulate symbols, but we probably don't have like a symbolic representation in our heads. Right. We have some distributed representation that is neural net, like in some way of lots of different neurons. And activation patterns firing when we see certain things and that enables us to reason and plan and, you know, do chains of thought and, you know, roll them back now that, that approach for solving the problem doesn't seem like it's going to work. I'm going to try this one. And, you know, in a lot of ways we're emulating what we intuitively think, uh, is happening inside real brains in neural net based models. So it never made sense to me to have like completely separate. Uh, discrete, uh, symbolic things, and then a completely different way of, of, uh, you know, thinking about those things.Shawn Wang [00:47:59]: Interesting. Yeah. Uh, I mean, it's maybe seems obvious to you, but it wasn't obvious to me a year ago. Yeah.Jeff Dean [00:48:06]: I mean, I do think like that IMO with, you know, translating to lean and using lean and then the next year and also a specialized geometry model. And then this year switching to a single unified model. That is roughly the production model with a little bit more inference budget, uh, is actually, you know, quite good because it shows you that the capabilities of that general model have improved dramatically and, and now you don't need the specialized model. This is actually sort of very similar to the 2013 to 16 era of machine learning, right? Like it used to be, people would train separate models for lots of different, each different problem, right? I have, I want to recognize street signs and something. So I train a street sign. Recognition recognition model, or I want to, you know, decode speech recognition. I have a speech model, right? I think now the era of unified models that do everything is really upon us. And the question is how well do those models generalize to new things they've never been asked to do and they're getting better and better.Shawn Wang [00:49:10]: And you don't need domain experts. Like one of my, uh, so I interviewed ETA who was on, who was on that team. Uh, and he was like, yeah, I, I don't know how they work. I don't know where the IMO competition was held. I don't know the rules of it. I just trained the models, the training models. Yeah. Yeah. And it's kind of interesting that like people with these, this like universal skill set of just like machine learning, you just give them data and give them enough compute and they can kind of tackle any task, which is the bitter lesson, I guess. I don't know. Yeah.Jeff Dean [00:49:39]: I mean, I think, uh, general models, uh, will win out over specialized ones in most cases.Shawn Wang [00:49:45]: Uh, so I want to push there a bit. I think there's one hole here, which is like, uh. There's this concept of like, uh, maybe capacity of a model, like abstractly a model can only contain the number of bits that it has. And, uh, and so it, you know, God knows like Gemini pro is like one to 10 trillion parameters. We don't know, but, uh, the Gemma models, for example, right? Like a lot of people want like the open source local models that are like that, that, that, and, and, uh, they have some knowledge, which is not necessary, right? Like they can't know everything like, like you have the. The luxury of you have the big model and big model should be able to capable of everything. But like when, when you're distilling and you're going down to the small models, you know, you're actually memorizing things that are not useful. Yeah. And so like, how do we, I guess, do we want to extract that? Can we, can we divorce knowledge from reasoning, you know?Jeff Dean [00:50:38]: Yeah. I mean, I think you do want the model to be most effective at reasoning if it can retrieve things, right? Because having the model devote precious parameter space. To remembering obscure facts that could be looked up is actually not the best use of that parameter space, right? Like you might prefer something that is more generally useful in more settings than this obscure fact that it has. Um, so I think that's always attention at the same time. You also don't want your model to be kind of completely detached from, you know, knowing stuff about the world, right? Like it's probably useful to know how long the golden gate be. Bridges just as a general sense of like how long are bridges, right? And, uh, it should have that kind of knowledge. It maybe doesn't need to know how long some teeny little bridge in some other more obscure part of the world is, but, uh, it does help it to have a fair bit of world knowledge and the bigger your model is, the more you can have. Uh, but I do think combining retrieval with sort of reasoning and making the model really good at doing multiple stages of retrieval. Yeah.Shawn Wang [00:51:49]: And reasoning through the intermediate retrieval results is going to be a, a pretty effective way of making the model seem much more capable, because if you think about, say, a personal Gemini, yeah, right?Jeff Dean [00:52:01]: Like we're not going to train Gemini on my email. Probably we'd rather have a single model that, uh, we can then use and use being able to retrieve from my email as a tool and have the model reason about it and retrieve from my photos or whatever, uh, and then make use of that and have multiple. Um, you know, uh, stages of interaction. that makes sense.Alessio Fanelli [00:52:24]: Do you think the vertical models are like, uh, interesting pursuit? Like when people are like, oh, we're building the best healthcare LLM, we're building the best law LLM, are those kind of like short-term stopgaps or?Jeff Dean [00:52:37]: No, I mean, I think, I think vertical models are interesting. Like you want them to start from a pretty good base model, but then you can sort of, uh, sort of viewing them, view them as enriching the data. Data distribution for that particular vertical domain for healthcare, say, um, we're probably not going to train or for say robotics. We're probably not going to train Gemini on all possible robotics data. We, you could train it on because we want it to have a balanced set of capabilities. Um, so we'll expose it to some robotics data, but if you're trying to build a really, really good robotics model, you're going to want to start with that and then train it on more robotics data. And then maybe that would. It's multilingual translation capability, but improve its robotics capabilities. And we're always making these kind of, uh, you know, trade-offs in the data mix that we train the base Gemini models on. You know, we'd love to include data from 200 more languages and as much data as we have for those languages, but that's going to displace some other capabilities of the model. It won't be as good at, um, you know, Pearl programming, you know, it'll still be good at Python programming. Cause we'll include it. Enough. Of that, but there's other long tail computer languages or coding capabilities that it may suffer on or multi, uh, multimodal reasoning capabilities may suffer. Cause we didn't get to expose it to as much data there, but it's really good at multilingual things. So I, I think some combination of specialized models, maybe more modular models. So it'd be nice to have the capability to have those 200 languages, plus this awesome robotics model, plus this awesome healthcare, uh, module that all can be knitted together to work in concert and called upon in different circumstances. Right? Like if I have a health related thing, then it should enable using this health module in conjunction with the main base model to be even better at those kinds of things. Yeah.Shawn Wang [00:54:36]: Installable knowledge. Yeah.Jeff Dean [00:54:37]: Right.Shawn Wang [00:54:38]: Just download as a, as a package.Jeff Dean [00:54:39]: And some of that installable stuff can come from retrieval, but some of it probably should come from preloaded training on, you know, uh, a hundred billion tokens or a trillion tokens of health data. Yeah.Shawn Wang [00:54:51]: And for listeners, I think, uh, I will highlight the Gemma three end paper where they, there was a little bit of that, I think. Yeah.Alessio Fanelli [00:54:56]: Yeah. I guess the question is like, how many billions of tokens do you need to outpace the frontier model improvements? You know, it's like, if I have to make this model better healthcare and the main. Gemini model is still improving. Do I need 50 billion tokens? Can I do it with a hundred, if I need a trillion healthcare tokens, it's like, they're probably not out there that you don't have, you know, I think that's really like the.Jeff Dean [00:55:21]: Well, I mean, I think healthcare is a particularly challenging domain, so there's a lot of healthcare data that, you know, we don't have access to appropriately, but there's a lot of, you know, uh, healthcare organizations that want to train models on their own data. That is not public healthcare data, uh, not public health. But public healthcare data. Um, so I think there are opportunities there to say, partner with a large healthcare organization and train models for their use that are going to be, you know, more bespoke, but probably, uh, might be better than a general model trained on say, public data. Yeah.Shawn Wang [00:55:58]: Yeah. I, I believe, uh, by the way, also this is like somewhat related to the language conversation. Uh, I think one of your, your favorite examples was you can put a low resource language in the context and it just learns. Yeah.Jeff Dean [00:56:09]: Oh, yeah, I think the example we used was Calamon, which is truly low resource because it's only spoken by, I think 120 people in the world and there's no written text.Shawn Wang [00:56:20]: So, yeah. So you can just do it that way. Just put it in the context. Yeah. Yeah. But I think your whole data set in the context, right.Jeff Dean [00:56:27]: If you, if you take a language like, uh, you know, Somali or something, there is a fair bit of Somali text in the world that, uh, or Ethiopian Amharic or something, um, you know, we probably. Yeah. Are not putting all the data from those languages into the Gemini based training. We put some of it, but if you put more of it, you'll improve the capabilities of those models.Shawn Wang [00:56:49]: Yeah.Jeff Dean [00:56:49]:

Heart disease remains the leading cause of death in the United States—but the good news is that many forms of heart disease are preventable with early detection, medical management, and lifestyle changes. On this week's Health Matters Radio Show, we sat down with Mihail T. Paxos, MD, cardiologist at Aultman Deuble Heart and Vascular Hospital, to discuss coronary artery disease (CAD) and what patients can do to protect their heart health. https://www.medshoprx.com/blog/coronary-artery-disease-amp-building-a-heart-healthy-lifestyle

Bhaumik (“Bhau”) Kotecha is the Co-Founder and Head of Paxos Labs, a startup spun out from Paxos in 2025 to empower institutions and developers with enterprise-grade infrastructure for on-chain financial products. At Paxos Labs, he leads the mission to remove the complexity of decentralized finance so partners can seamlessly embed trusted products—such as custom stablecoins, DeFi yield strategies, and tokenized assets—into their platforms. His vision blends the regulatory rigor and infrastructure expertise of Paxos with the agility and innovation of Web3, making it easier for financial services and fintech firms to adopt blockchain-based solutions at scale.Before launching Paxos Labs, Kotecha built his career at the intersection of product strategy and go-to-market execution. At Paxos, he led product strategy for the Tokenization team after driving growth, analytics, and operations for the firm. He previously spent nearly six years at Block (formerly Square), where he directed go-to-market efforts for Cash App and Square. Beyond his leadership at Paxos Labs, Kotecha is an active angel investor and advisor in the fintech and Web3 ecosystem, and he has recently spoken on the importance of open, developer-friendly stablecoin infrastructure—highlighting proposals like Paxos Labs' work on Hyperliquid's USDH stablecoin, which prioritizes accessibility, liquidity, and sustainable token economics.In this conversation, we discuss:- Difference between Paxos and Paxos Labs - Stablecoin adoption & liquidity - Infrastructure for custom-branded stablecoins - Tools for tokenized yield strategies & structured digital assets - The economics and interoperability of stablecoins - USDG and USDG0 - Off-platform dollar economics - Integrating yield, stablecoins, payments directly into existing apps - Enterprise grade security - Privacy-preserving stablecoins - PAXG - Paxos tokenized gold Paxos LabsX: @paxoslabsWebsite: www.paxoslabs.comLinkedIn: Paxos LabsBhaumik (“Bhau”) KotechaX: @bhau___LinkedIn: Bhaumik Kotecha---------------------------------------------------------------------------------This episode is brought to you by PrimeXBT.PrimeXBT offers a robust trading system for both beginners and professional traders that demand highly reliable market data and performance. Traders of all experience levels can easily design and customize layouts and widgets to best fit their trading style. PrimeXBT is always offering innovative products and professional trading conditions to all customers.  PrimeXBT is running an exclusive promotion for listeners of the podcast. After making your first deposit, 50% of that first deposit will be credited to your account as a bonus that can be used as additional collateral to open positions. Code: CRYPTONEWS50 This promotion is available for a month after activation. Click the link below: PrimeXBT x CRYPTONEWS50FollowApple PodcastsSpotifyAmazon MusicRSS Feed

Austin Campbell is a finance and risk management professional with two decades of experience spanning trading, portfolio management, executive leadership, and academia. He is the Managing Partner and Founder of Zero Knowledge Consulting and serves as an Acting CFO at Glueti, having recently held the role of Acting CEO at WSPN Ltd. He has taught as an adjunct professor at both NYU Stern and Columbia Business School, specializing in finance and markets. Previously, Austin was Chief Risk Officer and Head of Portfolio Management at Paxos, following senior trading and portfolio management roles at Citi, Stone Ridge, and JP Morgan Chase, where he advanced to Executive Director in Rates Trading. He began his career as a catastrophe risk analyst at Benfield and John B Collins Associates, with early research experience in mathematics at California State University Chico. In this conversation, we discuss:- Open Frontier - Stablecoins - Tokenization of assets - Traditional payment systems vs crypto - Decoupling lending incentives from user incentives - The importance of the Genius Act - Economic realignment that returns power to mainstream - The fragmentation of the financial systems - Composability of blockchains - “smart regulation” - Zero Knowledge Consulting Zero Knowledge Consulting X: @ZKZeroKnowledgeWebsite: www.zero-knowledge.comNewsletter: www.zero-in.beehiiv.comAustin CampbellX: @austincampbellLinkedIn: Austin Campbell---------------------------------------------------------------------------------This episode is brought to you by PrimeXBT.PrimeXBT offers a robust trading system for both beginners and professional traders that demand highly reliable market data and performance. Traders of all experience levels can easily design and customize layouts and widgets to best fit their trading style. PrimeXBT is always offering innovative products and professional trading conditions to all customers.  PrimeXBT is running an exclusive promotion for listeners of the podcast. After making your first deposit, 50% of that first deposit will be credited to your account as a bonus that can be used as additional collateral to open positions. Code: CRYPTONEWS50 This promotion is available for a month after activation. Click the link below: PrimeXBT x CRYPTONEWS50FollowApple PodcastsSpotifyAmazon MusicRSS FeedSee All

Join Alex Tapscott as he decodes the world of crypto with special guest Bhau Kotecha, Co-Founder and Head of Paxos Labs. Listen in as they discuss why stablecoins are becoming the default onchain financial building block, how Paxos and Paxos Labs power major tokenization and crypto brokerage initiatives (including PYUSD and USDG) while making DeFi services like lending, borrowing, staking, and yield strategies safer and easier for institutions to embed, why the Washington policy fight is shifting toward rewards and yield as the key battleground after the GENIUS Act, how stablecoins could reshape bank deposit economics while strengthening global dollar adoption, and why issuer strategy often starts on Ethereum before expanding cross-chain due to liquidity bootstrapping, developer network effects, and security considerations.

Nic Carter gooide de knuppel in het hoenderhok: quantum computing vormt een groter risico voor Bitcoin dan veel mensen denken. Maar klopt dat wel? We duiken in de discussie en geven je onze visie op het quantum-debat. Verder blikken we terug op een teleurstellend slot van 2025. Van een kerst-rally is geen sprake. De vraag is: wat nu?Probeer Bitcoin Alpha 2 weken gratis!Satoshi Radio wordt mede mogelijk gemaakt door: Firefish, Amdax, Watson Law en onze hoofdsponsor Bitvavo.Timestamps(00:00:00) Welkom en Podcast Introductie(00:08:00) Hebben de midterms in de VS impact op bitcoin?(00:22:00) Bookmark van Peter: Quantumdebat is weer losgebarsten(00:00:00) Bookmark van Bart: Wie wordt de everything exchange?(01:01:30) Bookmark van Bert: x402 adoption is taking off(01:07:00) Bookmark van Peter: De dubbele belangen van Tom Lee(01:10:00) Bookmark van Bart: Hut 8 Signs 15-Year, 245 MW AI Data Center Lease at River Bend Campus with Total Contract Value of $7.0 Billion(01:17:00) MarktupdateBookmarksBert:ETF's flow positive return negativex402 adoption is taking offBart:The future of finance is on CoinbaseRobinhood gaat ook prediction markets toevoegenSome people have already started to realize that using prediction markets can be cheaper than conventional fire, flood, and hurricane insuranceWhen currency is debased, everyone is a speculator. Hard work isn't enough, you have to "predict" the weather.Beleg in crypto bij jouw vertrouwde bankCircle, Ripple, Paxos, Fidelity and BitGo Get Banking Charters Approved by OCCHut 8 Signs 15-Year, 245 MW AI Data Center Lease at River Bend Campus with Total Contract Value of $7.0 BillionPartnership met AnthropicPeter:Quantumdebat is weer losgebarstenDe dubbele belangen van Tom Lee

Peter Schiff criticized CNBC for focusing on Bitcoin while overlooking substantial gains in gold and silver. At Binance Blockchain Week 2025, Binance founder CZ and Peter Schiff engaged in a fiery debate over the value and future of Bitcoin versus tokenized gold.~This episode is sponsored by iTrust Capital~iTrustCapital | Get $100 Funding Reward + No Monthly Fees when you sign up using our custom link! ➜ https://bit.ly/iTrustPaulGUEST: Peter Schiff. Founder of Schiff, Gold and Chief Economist and Global Strategist for Euro Pacific Asset ManagementSchiff's Tokeznied Gold ➜ https://bit.ly/SchiffTokenizedGold00:00 Intro00:10 Sponsor: iTrust Capital01:00 Won the debate with CZ?03:45 Peter Schiff was right?04:10 media only talks about crypto?06:40 Metals vs Crypto ETFs12:00 Tokenized Commodities market cap prediction?13:45 Is tokenized gold the future?15:00 Why not create a wallet (Schiff Card?)17:00 Gold bug psychology: will they ever trust tokenized gold?18:45 Is there a Silver cabal supply crisis hoax?21:50 Why isn't Schiff more bullish on Ethereum?23:30 Would you choose ETH for tokenized gold/silver?25:40 TradFi vs DeFi Gold?27:20 Should people trust their bank with their tokenized Gold?28:45 Outro#Gold #Crypto #Ethereum~Gold vs Tokenized Gold on Ethereum

Crypto News: THE OCC granted conditional approval for trust bank charters to Ripple, BitGo, Fidelity DigitalAssets, Paxos, and Circle. Wrapped XRP is coming to Solana for DeFi.Brought to you by

Bitcoin may be pitched as an alternative to the dollar system, but its price behavior shows how tightly it's now linked to the same forces that drive equities, credit, and tech multiples. When liquidity improves (when dollars are easier to borrow and funding markets relax), risk-taking becomes cheaper and more comfortable.~This episode is sponsored by iTrust Capital~iTrustCapital | Get $100 Funding Reward + No Monthly Fees when you sign up using our custom link! ➜ https://bit.ly/iTrustPaulGuest: Lyn Alden, Founder of Lyn Alden Investment StrategyLyn Alden website ➜ https://bit.ly/LynAldensiteBUY Lyn's Book "Broken Money" ➜ https://bit.ly/BrokenMoneyBook00:00 Intro00:10 Sponsor: iTrust Capital01:10 Debasement Trade03:00 Multi year QE04:30 Does Bitcoin need QE?05:30 Is the 4-Year cycle Dead?06:45 Michael Saylor $MSTR Strategy08:50 $BMNR outlook10:30 Interest rates goin to zero?12:40 Kevin Hassett as Fed Chair14:30 2 Years Ago: Lyn Alden Was Right vs VanEck15:30 Is Tokenization/Stablecoins taking away from Bitcoin narrative?16:45 Tokenized Gold vs Bitcoin Products19:40 Bitcoin All-time high in 2026?21:00 Would you add Paxos or Polymarket to your portfolio?22:15 Which AI stock/sector/options play are you considering?25:30 Is ZCash scam-pump finally over?25:20 Is now a good time to DCA?26:30 Outro#Crypto #Bitcoin #Ethereum~Market Liquidity Incoming

Crypto News: Texas becomes the FIRST state to purchase Bitcoin with a $5 million investment in BlackRock's BTC ETF IBIT. Klarna enters crypto with new USD stablecoin built on Stripe's Tempo chain. Kevin Hassett, who has crypto ties, rises to the front in Fed Chair search. Brought to you by ✅ VeChain is a versatile enterprise-grade L1 smart contract platform https://www.vechain.org/

Larry Wade, Global Head of Compliance and Regulatory Relations for Crypto at PayPal, joined me at Chainlink SmartCon to discuss PayPal's crypto services and its PYUSD stablecoin.Brought to you by ✅ VeChain is a versatile enterprise-grade L1 smart contract platform https://www.vechain.org/

For the first time since the government shutdown, new inflation data being released. Former Minneapolis Fed President Gary Stern breaks down the impact for future rate cuts. Then, Ford shares jumping on the back of results and plans to “significantly increase” its U.S. pickup truck production. Plus, the CEO of PayPal's crypto partner Paxos. His first TV interview since the company accidentally minted $300 trillion in stablecoins. He explains what happened and the impact on industry confidence. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

What if a “stable” coin wasn't? Blake and David unpack Paxos's accidental minting of $300 trillion in PayPal USD—and what it reveals about stablecoin “pegs.” They also dig into proposals to weaponize IRS Criminal Investigations, NASBA's flawed CPA exam data, and small-business headaches from forced AI rollouts in QuickBooks. Plus: Ripple's treasury move, whether AI can run a $1M solo firm, and smart guardrails for using AI without breaking your books.SponsorsOnPay - http://accountingpodcast.promo/onpayRelay - http://accountingpodcast.promo/relayBill.com - http://accountingpodcast.promo/billCloud Accountant Staffing - http://accountingpodcast.promo/casChapters(01:12) - Celebrating Spreadsheet Day (04:40) - Remote Work and Flexibility in Accounting (05:50) - Live Stream Interaction (06:16) - Political Discussion: IRS Overhaul (18:12) - NASBA Data Issues (24:51) - AI in Accounting: QuickBooks and Docket (41:28) - Upcoming Events and Conferences (42:38) - Navigating Conferences: Tips and Challenges (43:16) - The Future of Earmark: Audio Books and Discounts (46:21) - Chicago Teachers Union Audit Controversy (48:52) - Stablecoins and Cryptocurrency Concerns (56:09) - AI and the Future of Accounting (57:32) - The Talent Shortage in Accounting (01:03:30) - The CPA Exam and the Future of the Profession (01:16:48) - Government Shutdown and Its Impact (01:18:48) - Closing Remarks and Listener Engagement  Show NotesTrump Administration Plans IRS Overhaul to Target Left-Leaning Groups (Wall Street Journal) https://www.foxbusiness.com/politics/trump-admin-plans-irs-overhaul-probe-left-leaning-groups-more-easily-wsj-reportsNew CPA Success Index is DOA: NASBA's data behind the index no longer holds up https://www.accountingtoday.com/opinion/cpa-success-index-is-doa-nasbas-data-behind-the-index-no-longer-holds-upDocyt Launches Mission to Create the 'Million Dollar Accountant' https://www.cpapracticeadvisor.com/2025/10/15/docyt-launches-mission-to-create-the-million-dollar-accountant/170851/Ripple Acquires GTreasury for $1 Billion https://www.gtreasury.com/news/ripple-acquires-gtreasuryPaxos Accidentally Mints $300 Trillion 'Excess' PayPal Stablecoin https://www.pymnts.com/cryptocurrency/2025/paxos-accidentally-mints-300-trillion-excess-paypal-stablecoin/Dozens of Arkansas cities still without state funds over missing audits https://www.ktlo.com/2025/10/13/dozens-of-arkansas-cities-still-without-state-funds-over-missing-auditsChicago Teachers Union members sue union bosses over missing financial reports https://www.illinoispolicy.org/chicago-teachers-union-members-sue-union-bosses-over-missing-financial-reports/Need CPE?Get CPE for listening to podcasts with Earmark: https://earmarkcpe.comSubscribe to the Earmark Podcast: https://podcast.earmarkcpe.comGet in TouchThanks for listening and the great reviews! We appreciate you! Follow and tweet @BlakeTOliver and @DavidLeary. Find us on Facebook and Instagram. If you like what you hear, please do us a favor and write a review on Apple Podcasts or Podchaser. Call us and leave a voicemail; maybe we'll play it on the show. DIAL (202) 695-1040.SponsorshipsAre you interested in sponsoring The Accounting Podcast? For details, read the prospectus.Need Accounting Conference Info? Check out our new website - accountingconferences.comLimited edition shirts, stickers, and other necessitiesTeePublic Store: http://cloudacctpod.link/merchSubscribeApple Podcasts: http://cloudacctpod.link/ApplePodcastsYouTube: https://www.youtube.com/@TheAccountingPodcastSpotify: http://cloudacctpod.link/SpotifyPodchaser: http://cloudacctpod.link/podchaserStitcher: http://cloudacctpod.link/StitcherOvercast: http://cloudacctpod.link/OvercastClassifiedsCollective by DBA - https://collective.cpa/ Want to get the word out about your newsletter, webinar, party, Facebook group, podcast, e-book, job posting, or that fancy Excel macro you just created? Let the listeners of The Accounting Podcast know by running a classified ad. Go here to create your classified ad: https://cloudacctpod.link/RunClassifiedAdTranscriptsThe full transcript for this episode is available by clicking on the Transcript tab at the top of this page

The day after last week's show, Apple announced its new M5 chip alongside a new Vision Pro, MacBook Pro, and iPad Pro. AWS had a major outage that took down good chunks of the internet. The Louvre had horrible security. All that and so much more to cover this week, so you can get out there and tech better. Watch on YouTube! - Notnerd.com and Notpicks.com INTRO (00:00) MAIN TOPIC: Apple M5 and F1 (05:40) Apple announces M5 chip and new Vision Pro, MacBook Pro, iPad Pro Apple and Formula 1 ink 5 year exclusive U.S. streaming deal worth about $750 million Apple and NBCUniversal introduce the Apple TV and Peacock Bundle Apple's stock price reaches new all-time high DAVE'S PRO-TIP OF THE WEEK: Get a Callback Reminder for a Missed Call (21:05) JUST THE HEADLINES: (24:45) Paxos mistakenly issues $300 Trillion of PayPal Stablecoin, exceeding global currency supply Steve Jobs to be honored on 2026 American Innovation $1 Coin Charmin announces new toilet paper roll designed to last up to a month Kohler unveils a camera for your toilet DOJ seizes $15 billion in bitcoin from massive ‘pig butchering' scam based in Cambodia Laywer caught using AI while explaining to court why he used AI The Numbers Six and Seven Are Making Life Hell for Math Teachers LISTENER MAIL: Don't use your work email for personal stuff PSA (31:00) TAKES: AWS services recover after daylong outage hits major sites (33:10) Veritas AI program for high schools (38:55) Louvre heist raises decades old questions about museum security - How the Louvre jewelry heist unfolded (41:45) BONUS ODD TAKE: 50 unedited photos so confusing they don't look real (45:30) PICKS OF THE WEEK:  Dave:  Amazon Fire HD 10 tablet (newest model) built for relaxation, 10.1" vibrant Full HD screen, octa-core processor, 3 GB RAM, 32 GB, Ocean (49:15) Nate: NEEWER Selfie Stick with Detachable Bluetooth Remote Shutter, Lightweight Foldable Travel Tripod for iPhone for YouTube/TikTok Vlogs Live Streaming Photography, Black, P15 (58:00) RAMAZON PURCHASE OF THE WEEK (01:01:35)

$300 Trillion Stablecoin Mint Mistake | Paxos & PYUSD | Is the World Ready for Stablecoins? Paxos, the blockchain partner behind PayPal's PYUSD, just made headlines with a shocking $300 trillion minting mistake — an accidental creation of more tokens than the global economy itself. In tonight's episode of On The Chain, Jeff and Chip break down:

Esta semana en Espacio Cripto, analizamos si estamos viendo el final del bull market tras la caída de Bitcoin, que volvió a los $107,000 USD, y Ethereum bajando más del 6%. Pero entre tanta volatilidad, hay señales muy bullish: Sony, Charles Schwab y Citi están entrando al mundo cripto con fuerza.Inscríbete a Inversionista del Futuro: https://www.espaciocripto.io/inversionista?utm_source=social&utm_medium=yt&utm_content=bioComunidad de Espacio Cripto: https://t.me/espaciocripto0:00 Intro1:45 ¿Se acabó el bull market? 4:30 Estrategias personales para el bear market 6:30 Precio actual de Bitcoin, Ethereum, Solana y BNB 7:20 Sony lanza su propio banco cripto 8:00 Charles Schwab planea ofrecer Bitcoin y trading cripto 11:30 Bridge busca licencia bancaria en EE.UU. 13:30 Bancos y exchanges: la nueva competencia 14:30 Error de Paxos: 300 trillones minteados por error 15:40 El oro alcanza máximos históricos 18:00 Por qué invertir en oro vs altcoins 18:50 Uniswap llega a Solana 19:50 Tempo levanta $500M respaldada por Stripe 21:00 El airdrop de Monad y su enfoque en builders

Markets just saw a $19B wipeout in a single day. In this week's Weekly Rollup, we break down the Friday Flash Crash, what really caused it, and whether it signals the end of the cycle or just a reset. We also cover Binance's leaked listing fees, a major Chinese tech company quietly building on Ethereum, and reports that the U.S. may add $14B in Bitcoin to its strategic reserve. Plus, Democrats renew their push against DeFi, and Larry Fink gears up for BlackRock's next big crypto play. ------

Matt and Nic are back with another week of news and deals. In this episode:  We dissect the market structure behind the flash crash last week What happened with Binance and Ethena? Palmer Luckey's de novo bank Erebor gets an OCC charter Paxos' fat finger The US government has seized $15b in funds tied to pig butchering scams  

Mike Armstrong and Paul Lane discuss explain why people shouldn't bo overly surprised that markets have so far defied inflation and tariffs. Banks' strong earnings leave investors digging deeper for trouble spots. Small businesses are being crushed by Trump's tariffs and economists say it's a warning for the economy. Honda wants you to teach its new robot lawnmower. Paul LaMonica (Barron's) joins the show to chat about PayPal getting $300T worth of  bad news from Paxos.

Today's blockchain and cryptocurrency news  Paxos mistakenly mints $300 trillion PYUSD Bitcoin whale selling and put demand intensify Kraken buys Small Exchange for $100 million to launch US derivatives trading platform Australia moves to target crypto ATMs under new AML powers BitMine adds $417 million worth of Ethereum ###Gemini Card Disclosure: The Gemini Credit Card is issued by WebBank. In order to qualify for the $200 crypto intro onus, you must spend $3,000 in your first 90 days. Terms Apply. Some exclusions apply to instant rewards in which rewards are deposited when the transaction posts. This content is not investment advice and trading crypto involves risk. For more details on rates, fees, and other cost information, see Rates & Fees. The Gemini Credit Card may not be used to make gambling-related purchases.  Learn more about your ad choices. Visit megaphone.fm/adchoices

Paxos mistakenly minted $300 trillion of PYUSD on Ethereum. Stablecoin issuer Paxos accidentally minted $300 trillion worth of PayPal's PYUSD stablecoin, far exceeding the total U.S. dollar supply. Paxos emphasized that it was not a security breach but the incident raised concerns about how an enormous amount of stablecoin could be created without requisite collateral. CoinDesk's Jennifer Sanasie hosts "CoinDesk Daily." - Break the cycle of exploitation. Break down the barriers to truth. Break into the next generation of privacy. Break Free. Free to scroll without being monetized. Free from censorship. Freedom without fear. We deserve more when it comes to privacy. Experience the next generation of blockchain that is private and inclusive by design. Break free with Midnight, visit midnight.network/break-free - Bridge simplifies global money movement. As the leading stablecoin issuance and orchestration platform, Bridge abstracts away blockchain complexity so businesses can seamlessly move between fiat and stablecoins. From payroll providers and remittance companies to neobanks and treasury teams, Bridge powers payments, savings, and stablecoin issuance for thousands – like Shopify, Metamask, Remitly, and more. URL: ⁠⁠⁠⁠⁠⁠⁠https://hubs.ly/Q03KGbRK0⁠⁠⁠⁠⁠⁠⁠ - OwlTing (Nasdaq: OWLS) is building invisible rails for global payments. With OwlPay, businesses and users can bridge fiat and stablecoins, send money instantly across borders, and access stablecoin checkout at lower costs. Licensed worldwide, OwlTing delivers secure, compliant, and regulated infrastructure for the digital economy. Learn more at owlting.com. - This episode was hosted by Jennifer Sanasie. “CoinDesk Daily” is produced by Jennifer Sanasie and edited by Victor Chen.

This live session is a Crypto Town Hall discussion with an expert panel discussing recent crypto industry events, market cycles, and broader financial trends. The goal is to break down current issues affecting crypto—from stablecoin mishaps and global scams to market cycles, the role of central banks, and the interplay between gold, Bitcoin, stocks, and treasuries. The conversation captures market sentiment, regulatory concerns, and evolving narratives, offering both macro views and ground-level insights for traders and investors.

De podcast stond amper online en het ging fout. Op vrijdagavond dook de hele cryptomarkt diep in het rood. Bitcoin ging 16% omlaag, sommige altcoins gingen letterlijk naar nul. In deze uitzending staan we er uitgebreid bij stil. Wat veroorzaakte de crash? En welke invloed hebben perpetual futures of de spot markt? Bert legt het je uit. Voor de rest hebben we een aantal leuke luisteraarsvragen en bookmarks. Zo hebben we het over intrinsieke waarde, goud, een enorme bitcoin vangst in Amerika en een foutje van Paxos. Veel luisterplezier!Probeer Bitcoin Alpha 2 weken gratis!Satoshi Radio wordt mede mogelijk gemaakt door: Amdax, Watson Law en onze hoofdsponsor Bitvavo.Timestamps(00:00:00) Welkom en Podcast Introductie(00:07:00) De crash van vorige week vrijdag(00:28:00) Vragen van luisteraars: Lagarde en Goud(00:40:00) Vragen van luisteraars: geld stallen in Zwitserland?(00:55:00) Bookmark van Bert: Luxembourg's Intergenerational Sovereign Wealth Fund (FSIL) has invested 1% of its holdings in Bitcoin(01:01:00) Bookmark van Bart: Grootste Bitcoin vangst voor US Department of Justice(01:09:00) Bookmark van Peter: De rol van Binance in de crash van afgelopen vrijdag(01:24:00) Bookmark van Bart: Back in the earlier Bitcoin days (2013 for me), we often thought of altcoins as a sort of testnet.(01:33:00) Bookmark van Peter: Paxos print $300.000.000.000.000(01:35:00) Bookmark van Bert: Five takeaways from the Bitwise Invest Crypto Diligence Summit(01:40:00) Marktupdate(02:32:00) EindeBookmarksBert:Luxembourg's Intergenerational Sovereign Wealth Fund (FSIL) has invested 1% of its holdings in BitcoinFive takeaways from the Bitwise Invest Crypto Diligence SummitBart:Grootste Bitcoin vangst voor US Department of JusticeTime to do it again.Zijn we eindelijk veilig voor ‘Chat Control'?Back in the earlier Bitcoin days (2013 for me), we often thought of altcoins as a sort of testnet.Peter:Binance - Crashtheorie 1: Het was een aanvalBinance - Crashtheorie 2: Het was Binance zelfBinance - Angry mob 1: ‘Haal er je assets weg'Binance - Angry mob 2: ‘Sjoemelt met token listings'Binance - ‘Chinese PR in een Westerse community'Larry Fink wordt blij van tokenizationPaxos print $300.000.000.000.000

Stablecoins are no longer a side story — they're on the path to becoming the backbone of global digital finance.To unpack what the GENIUS Act means for the U.S. dollar, stablecoin issuers, and banking competition, I sat down with Austin Campbell, Founder and Managing Partner of Zero Knowledge Consulting and an Adjunct Professor at Columbia Business School.Austin previously led Stable Value Trading at JP Morgan, co-headed Digital Asset Rates Trading at Citi, and served as Head of Portfolio Management at Paxos. In this episode, Austin explains the key provisions of the Genius Act, the misconceptions around the “interest” prohibition, and how competition between currencies could expand freedom — and reshape the global economy.Timestamps:➡️ 00:00 — Intro➡️ 00:46 — Sponsor: Day One Law➡️ 01:09 — Austin's path from Wall Street to crypto➡️ 05:40 — Why the Genius Act is the most important bipartisan financial law since Dodd-Frank➡️ 10:31 — Stablecoins as global infrastructure for the U.S. dollar➡️ 15:14 — Key pillars of the Genius Act: reserves, insolvency, and compliance➡️ 26:20 — Privacy, enforcement, and what Genius gets right➡️ 37:19 — The “interest” prohibition — and the exception most people missed➡️ 45:00 — What comes next for stablecoin issuers and U.S. regulators& much more.Sponsor: ⁠⁠ This episode is brought to you by ⁠⁠Day One Law, a boutique law firm helping crypto startups navigate complex legal challenges. Subscribe to ⁠⁠Day One's free monthly newsletter⁠⁠ for legal and regulatory updates.Resources:

Stablecoins are having a moment. In this episode of TRM Talks, Ari sits down with Lesley Chavkin, Head of Global Policy at Paxos and former US Treasury official, to unpack the regulatory evolution shaping the future of digital finance.Lesley shares her journey from building anti-money laundering frameworks in the Middle East, to building at Stellar, to now helping guide Paxos through a pivotal period for stablecoins. Her perspective bridges public service and private sector innovation — making her one of the most thoughtful voices in crypto policy today.The conversation covers the impact of the GENIUS Act and why it marks a turning point for stablecoin issuers, how Paxos' regulatory-first model is resonating with banks and traditional finance, and the differences between regimes like Singapore, the EU's MiCA, and the US. They also dive into what financial institutions need to know about stablecoin infrastructure — from partnerships with PayPal to the buildout of the Global Dollar Network.Lesley offers a candid view on what policymakers often miss about digital assets — and what's still needed to build a safe, trusted, and interoperable future for stablecoins.

Welcome to The Chopping Block – where crypto insiders Haseeb Qureshi, Tom Schmidt, Tarun Chitra, and Robert Leshner chop it up about the latest in crypto. This week, we're joined by Gordon Liao, Chief Economist at Circle, to dissect the Stablecoin Wars. From Circle's Arc and Stripe + Paradigm's Tempo, to Solana's native stablecoin push and Hyperliquid's deal, we unpack why everyone suddenly wants their own chain or branded stablecoin. Is this the future of crypto's monetary layer — or just a fragmentation nightmare? We dig into FX use cases, PMF for stablecoins, collective bargaining power of ecosystems, and whether “stablecoin-as-a-service” is the next killer primitive or a liquidity trap. Show highlights

Austin Campbell, Managing Partner at Zero Knowledge Consulting, and Adam Zarazinski, CEO at Inca Digital, explore the Genius Bill and its potential to revolutionize the stablecoin landscape. They discuss the many changes with the current government, including regulations and compliance, along with how this new legislative proposal could drive innovation and widespread adoption of stablecoins. Key Takeaways:  Why the Genius Bill matters, and its potential impact on stablecoin adoption New compliance experiments and why the traditional approach is dead The brain drain of crypto experts within banking The need for innovation and technological transformation in traditional finance Guest Bio: Austin Campbell is the managing partner and founder of Zero Knowledge Consulting, and an adjunct professor at Columbia Business School. Previously, he ran the stablecoin platform and portfolio management at Paxos, and was the chief risk officer of Paxos National Trust. Austin has managed fixed income trading desks at JP Morgan and Citi covering >$100B of cash derivatives, and has been a portfolio manager and structurer at Stone Ridge, the parent of NYDIG. He holds a BS in Mathematics from CSU Chico, and an MBA from NYU Stern. Adam Zarazinski is the CEO of Inca Digital, a fintech company that provides data and analytics on the digital asset ecosystem to government agencies and financial institutions using natural language processing and LLMs. ---------------------------------------------------------------------------------------- About this Show: The Brave Technologist is here to shed light on the opportunities and challenges of emerging tech. To make it digestible, less scary, and more approachable for all! Join us as we embark on a mission to demystify artificial intelligence, challenge the status quo, and empower everyday people to embrace the digital revolution. Whether you're a tech enthusiast, a curious mind, or an industry professional, this podcast invites you to join the conversation and explore the future of AI together. The Brave Technologist Podcast is hosted by Luke Mulks, VP Business Operations at Brave Software—makers of the privacy-respecting Brave browser and Search engine, and now powering AI everywhere with the Brave Search API. Music by: Ari Dvorin Produced by: Sam Laliberte  

914INC Magazine's Best of Business celebration took place on Thursday, September 11, at the Royal Regency Hotel in Yonkers, honoring the exceptional accomplishments of Westchester's vibrant business community. The event gathered local business leaders and professionals for an evening of networking and recognition, spotlighting the top companies across a variety of B2B categories. Attendees enjoyed the chance to connect with peers, exchange ideas, and celebrate success stories that drive the region's economy. Westchester Talk Radio host Bob Marrone added an engaging touch to the evening, interviewing key figures including Maria Paxos-Pampafikos, executive vice president of the Royal Regency Hotel, who welcomed guests and helped make the night memorable.

Charles Cascarilla, CEO and Co-Founder of Paxos, joined me to discuss the impact of stablecoin legislation on the crypto market and what the future of payments looks like with stablecoins.Topics:- GENIUS Act passing impact on the Stablecoin market - Paxos application to convert NYDFS trust charter into a national trust charter under the OCC - Global Dollar Network (USDG) - PayPal's PYUSD- Tokenization market - Future of payments - CLARITY Act Crypto market structure- Will Paxos go public soon? Show Sponsor -

HyperLiquid 通过「比武招亲」的方式选定全新生态稳定币「USDH」，这让我们可以借这个机会和大家聊一聊，除了 USDT 和 USDC 之外，加密世界主要的二线稳定币还有哪些，以及他们是怎么运作的？ 另外，这也是观察稳定币真正的竞争力从何而来的绝佳案例。 本期节目发布时，HyperLiquid 社区横空出世的 Native Markets 宣布已经赢得 「USDH」 的发行权。 【主播】 刘锋，BODL Ventures 合伙人，前链闻总编辑 熊浩珺Jack，律动 BlockBeats 副主编，《Web3 无名说》主播 【嘉宾】 杨民道，DForce创始人 【赞助商】 本期节目由开源硬件钱包OneKey赞助播出。硬件钱包是保护加密资产最有效的方式之一。加密世界是黑暗森林，每个人都该为自己的资产负责。物理隔离的硬件钱包，能更好保护助记词不被盗窃。 访问OneKey官网 (https://shop.onekey.so/discount/web3101)购买开源硬件钱包，使用折扣码「web3101」可享受95折优惠。 【硅谷101科技峰会】 《硅谷101》的年度科技大会又回来了，这是我们将有趣的技术干货与故事带到线下的第二年，用最好的内容让大家亲身感受前沿科技的酷炫和温度一直是硅谷101的目标。硅谷10月5日，期待与各位见面，从这里驶向未来！ 欢迎点击前往 (https://luma.com/mtqq79ii)购票地址，输入粉丝专属折扣码“VALLEY101FANS”，享85折优惠。 【你将听到】 ** **HyperLiquid 稳定币招商大会 00:09 「招商大会」背景 03:28 HyperLiquid 用 USDC 等于每年向竞争对手 Coinbase 送出 1 亿美元 09:38 主要 6 家投标稳定币可分为两类 细数竞标稳定币 10:07 Paxos：主打合规，是稳定币界的富士康 14:35 Agora：新晋选手，主打传统金融与 DeFi 的结合 19:01 Native Markets：从 HyperLiquid 社区横空出世，胜算最大 27:40 Frax：老牌 DeFi degen，多次转型，日渐式微 33:24 Ethena：本轮周期现象级稳定币，增长迅速，但与 HyperLiquid 存在竞争 44:57 Sky：曾经的 MakerDAO，与 USDC 无损互换是最大优势 观察和感悟 51:00 这场招商对生态发展意义重大 52:34 中心化交易所会心有戚戚焉 53:57 稳定币的生意：发行为王 【名词解释】 本期提到的 Web3 词汇： Degen：DeFi 圈子非常流行流行的「黑话」，形容高风险偏好者，也指参与高风险的行为，追逐超高收益、参与早期和实验性项目，充当DeFi 领域的「先锋」。这个词带一定自嘲色彩，已经成为了一个圈内共识的文化标签。 【后期】 AMEI 【BGM】 Mumbai - Ooyy First Horizon - ELFL 【在这里找到我们】 收听渠道：苹果｜小宇宙 海外用户：Apple Podcast｜Spotify｜Google Podcast｜Amazon Music 联系我们：podcast@sv101.net

Hyperliquid's USDH ticker set off the most dramatic “RFP” in recent memory. The crew breaks down why Native Markets ran away with validator support, whether the process was theater or strategy, and how the Bake-off became a marketing masterstroke—and potential leverage on Circle. We dig into Polymarket odds, the last‑minute Paxos bribery allegation (denied), and what this means for future “native” stables on Solana, app chains, and beyond. Welcome to The Chopping Block – where crypto insiders Haseeb Qureshi, Tom Schmidt, Tarun Chitra, and Robert Leshner chop it up about the latest in crypto. This week, we're joined by Guy founder of Ethena as a special guest, as a single ticker (USDH) sparked a weeklong spectacle: Hyperliquid's “Bake-off” to award the USDH stablecoin brand. Native Markets surged ahead as validators signaled support, Paxos rallied late with partners and incentives, and Ethena ultimately withdrew. Was this always a vibes‑based beauty contest, or a deliberate move to pressure Circle and re‑route bridge yield? We parse the incentives, the governance, and the market microstructure — and peek at what happens if every big chain/app tries the “native stablecoin” playbook. Show highlights

Stablecoins are supposed to be boring, but the fight for the USDH ticker on Hyperliquid has turned into one of the most dramatic battles in crypto.  From Ethena suddenly pulling out, to Paxos revamping its bid, to whispers of favoritism, the contest has put protocol-native stablecoins in the spotlight.  In this episode, MegaETH co-founder Shuyao Kong, who just announced their own stablecoin USDm, speaks about why they chose Ethena as a partner, and why alignment with Hyperliquid matters more than short-term incentives.  She also explains why ecosystems need both yield-chasing and yield-agnostic stablecoins — and whether Circle and Tether could be pushed aside in the next wave of competition. Thank you to our sponsor, ⁠Token2049⁠! Get 15% off your tickets with the code UNCHAINED! Guest: Shuyao Kong, Co-founder of MegaETH Links:  Unchained:  The Competition Is On. Who'll Win the USDH Ticker on Hyperliquid? Bits + Bips: Hyperliquid's USDH Bidding War & Why the DAT Model Is Broken Ethena Joins Race to Issue Hyperliquid's USDH Paxos Unveils USDH Proposal V2 With PayPal, Venmo Integrations  Sky Joins Bidding War to Launch Hyperliquid's USDH Stablecoin Issuers Enter Bidding War to Launch Hyperliquid's USDH Ethena withdraws its proposal  The Block: MegaETH launches native USDm stablecoin with Ethena to subsidize sequencer fees Polymarket bet: Who will win the USDH ticker? Timestamps:

The battle for Hyperliquid's USDH ticker is a referendum on what crypto wants to be: a community-first public good, or a globally scaled, institution-ready product.  With the clock ticking toward the proposal and voting deadlines, Agora's Nick van Eck and Paxos Labs' Bhau Kotecha lay out their cases—100% revenue back to users vs. 95% with enterprise distribution, “Hyperliquid alignment” versus “bring it to the masses,” and what each would build on day one if they win.  We also dig into liquidity, slippage claims, validator dynamics, and how a single ticker could shape the future of onchain markets. Thank you to our sponsors ⁠⁠Mantle⁠⁠! Guests: Nick van Eck, CEO and Co-founder of Agora Bhau Kotecha, Co-founder of Paxos Labs Links: Unchained:  Stablecoin Issuers Enter Bidding War to Launch Hyperliquid's USDH Sky Joins Bidding War to Launch Hyperliquid's USDH  Timestamps:

Gold surged past the $3,600 an ounce level for the first time on Monday, hitting a record high, as soft U.S. labor data reinforced expectations the U.S. Federal Reserve will cut interest rates next week. Meanwhile, tokenized gold continues to rocket on Ethereum.~This episode is sponsored by iTrust Capital & Gemini~iTrustCapital | Get $100 Funding Reward + No Monthly Fees when you sign up using our custom link! ➜ https://bit.ly/iTrustPaulSign up for The Gemini Credit Card and get an extra $50 in crypto!➜ https://bit.ly/GeminiPBN00:00 Intro00:10 Sponsor: iTrust Capital00:40 Gold all time high01:10 Gold is the safe Bitcoin01:30 El Salvador buys gold over Bitcoin01:50 Tether dumping Bitcoin for gold02:10 Gold vs Treasuries02:30 Gold mindshare vs Bitcoin03:00 It's Already happening in Venezuela04:15 Onchain gold is cheaper04:45 ETH Gold up 100% YoY05:00 Ethereum gold rush to DeFi05:20 Gold yields on telegram06:10 Asia - Hong Kong Gold06:50 Sui Gold07:00 World Gold Council 202607:30 Federal reserve independence fears08:00 Sponsor: Gemini08:40 Peter Schiff was right09:45 Stocks & collectibles on ethereum11:30 Digital gold has cultural roots for kids too12:20 The Flip12:40 Outro#Crypto #Ethereum #Gold~Ethereum Boosting Gold To New All-Time Highs!

The stablecoin wars have arrived at HyperLiquid's doorstep.In today's episode, we dive into the USDA showdown that took over crypto Twitter this weekend. HyperLiquid opened up their native stablecoin ticker to community proposals, triggering a fierce competition between major players.We sit down with three key contenders: Nick from Agora, Bo from Paxos Labs, and Sam from Frax. Each brings their own vision for scaling USDA beyond HyperLiquid's $5.5 billion ecosystem - from institutional partnerships to DeFi-native coalition building.This isn't just about a ticker. It's about who becomes the steward of flows for one of crypto's most important trading platforms, with proposals offering 95-100% revenue sharing back to the community.---

In this episode, John Byrne and Elliot Berman unpack a series of significant developments in banking, compliance, and enforcement. They begin with the White House's new executive order on “Guaranteeing Fair Banking for All Americans,” which aims to prevent what some call “debanking.” While positioned as a fairness measure, John and Elliot warn that it could weaken banks' ability to make independent, risk-based decisions, potentially increasing white-collar crime exposure. They note concerns over the subjective nature of risk scoring and parallels to the 2008–2010 financial crisis. Next, they spotlight the IRS-CI “CI-FIRST” program, a collaborative effort between financial institutions and the IRS's Criminal Investigation division to improve information sharing and streamline financial record requests. The recent CI-FIRST Executive Forum in Washington is seen as a model for effective public-private partnerships in combating financial crime. The discussion then turns to enforcement actions: Paxos Trust Company will pay $26.5 million to New York regulators for failing to properly vet Binance and for systemic AML program weaknesses, alongside a $22 million investment in compliance upgrades. The DOJ issued its first corporate FCPA action since resuming enforcement, with Liberty Mutual paying $4.7 million to resolve bribery allegations involving Indian state-owned banks. Do Kwon, co-founder of Terraform Labs, pled guilty to wire fraud and conspiracy, tied to the $40 billion collapse of Terra USD and Luna, with a $19 million penalty and possible 12-year sentence. They also cover a Senate minority report critical of the administration's approach to Russian sanctions, arguing it undermines Ukraine's leverage and lacks consistent enforcement. The FACT Coalition emphasizes the need for tools like the Corporate Transparency Act to bolster sanctions' effectiveness. On the policy front, they discuss delays and staffing cuts affecting the State Department's annual human rights report and the pending trafficking in persons report—both key references for global human rights and anti-trafficking efforts.

Crypto News: Jeff Bezos's Blue Origin will accept crypto payments for space flights. Asset include Bitcoin, Ethereum Solana, USDT, USDC. Chainlink partners with the Intercontinental Exchange. Paxos joins Ripple and Circle in pursuit of seeking national bank charter licenses.Show Sponsor -

Gold prices fell on Monday as investors awaited White House clarification regarding potential U.S. tariffs on imported gold bars as well as a U.S. inflation report that could provide an indication of the Federal Reserve's rate outlook.~This episode is sponsored by iTrust Capital~iTrustCapital | Get $100 Funding Reward + No Monthly Fees when you sign up using our custom link! ➜ https://bit.ly/iTrustPaulGuest: Andy Schectman | President & Owner of Miles Franklin Miles Franklin website ➜ https://milesfranklin.com/Miles Franklin Youtube channel ➜ https://www.youtube.com/@MilesFranklinMedia00:00 Intro00:06 Sponsor: iTrust Capital00:35 Trump Gold Tariff Chaos!01:55 Tariff Whiplash Exposes Gold Market06:19 Gold Withdrawal Delays09:48 Uncertainty Bad Enough10:13 Tokenized Gold Explodes12:49 BRINKS Stock Pumps13:20 TradFi Moving to Tokenization for Trust15:15 Paxos & Tether Gold Going Cross-Chain16:10 Gold DeFi Yields19:10 Katana DeFi Yields on Polygon20:04 Are Gold Yields a deal-breaker?21:15 Tokenized Gold Mines22:15 Gold DeFi Business Models23:14 Gold Bugs trusting crypto more?24:47 Gold-Backed Stablecoins vs BRICS27:27 Miles Franklin Update28:50 outro#Crypto #Gold #Ethereum~Gold Tariffs!? vs Crypto

The balance between cock-ups and funny bits is a delicate one - we hope we've got it right... In this episode, Jane and Fi chat Mariah Carey's denial of time, Paxo and Paxos, the future of allotments, and Blue Tits.Plus, actor Adrian Dunbar, famous for Line of Duty, discusses the new season of 'Ridley'. The bag scheme Fi was talking about is here: https://www.madlug.com/?srsltid=AfmBOophSKkob9HecUQVYtaqe0e_U322WQ9QBbX9WKrf0tnX3TLUrVDO You can listen to the playlist here: https://open.spotify.com/playlist/3qIjhtS9sprg864IXC96he?si=uOzz4UYZRc2nFOP8FV_1jg&pi=BGoacntaS_ukiIf you want to contact the show to ask a question and get involved in the conversation then please email us: janeandfi@times.radioFollow us on Instagram! @janeandfiPodcast Producer: Eve SalusburyExecutive Producer: Rosie Cutler Hosted on Acast. See acast.com/privacy for more information.

In this episode, Nicolas Cary, Co-Founder of Blockchain.com, is joined by Lesley Chavkin, Global Head of Public Policy at Paxos, for a deep dive into PAX Gold (PAXG) - a regulated, asset-backed token representing one fine troy ounce of gold. They explore how PAXG fits into Paxos' broader tokenization strategy, what sets it apart from traditional gold exposure, and how it's being used by traders, institutions, and regulators alike.From real-world utility to the future of tokenized commodities, this is your all-access look at how blockchain is modernizing gold.

Tune in to this episode of the Security Token Show where this week Herwig Konings and guest contributors, Tal Elyashiv, Managing Partner at SPiCE VC, Nico Pantelis, Partner at Blue Bay Ventures, and Jason Barraza cover the industry leading headlines and market movements, including Robinhood's tokenized stocks, stablecoin issuers like Circle and Ripple applying for bank charters, SPiCE VC's 3rd investor payout, and more RWA news!   This week Jason Barraza had a chance to sit with Yuval Rooz, Co-Founder and CEO of Digital Asset, for an inside look at their $135 million raise, the growing momentum behind Canton Network including integrating with Nasdaq Calyspo, and why privacy, composability, and real world asset tokenization as a whole are going to change global markets.  Did Yuval leave you on a cliff hanger before the RWA Foundation Updates? Jason and Yuval dive into it and more on the full interview available here: https://youtu.be/Aj4URNZo9uM    Company of the Week - Herwig: Robinhood   Companies in the news include SPiCE VC, Circle, Ripple, Robinhood, Republic, SpaceX, OpenAI, S&P Dow Jones Indices, Centrifuge, Anemoy, Janus Henderson, Midas, Fasanara, Morpho, Steakhouse Financial, Bitfinex Securities, Castle Community Bank, Arbra, Colb, Gemini, Arbitrum, Euroclear, Abu Dhabi Securities Exchange, FAB, HSBC, Bybit, xStocks, Dinari, BitGo, Spiko, Chainlink, MetaWealth, Securitize, Redstone, Apex Group, GLEIF, ERC-3643 Association, Kinexys by J.P.Morgan, S&P Global, Particula, 21X, GF Securities, Hashkey, OnRe, Lynq Network, European Central Bank, Ondo, Pantera, Zult, Stable, AMINA, Ripple, Paxos, World Liberty Financial, Plume, Paul Atkins/ SEC, SIFMA   Check Out Tal's Book “Investing in Revolutions”: https://www.amazon.com/Investing-Revolutions-Creating-Transformational-Technology/dp/B0DKS3YFWV#detailBullets_feature_div    TokenizeThis 2025 Conference Review: https://docsend.com/v/k8bn7/tt25  STM Predicts $30-50T in RWAs by 2030: https://docsend.com/view/7jx2nsjq6dsun2b9    More STM.co Reports: https://reports.stm.co/  Join the RWA Foundation and Read the Whitepaper: RWAF.xyz  Learn More About WALLY DAO: WallyDAO.xyz    ⏰ TABLE OF CONTENTS ⏰ 0:00 Introduction 0:16 Welcome 0:47 Market Movements 34:06 STS Interviews: Yuval Rooz, Digital Asset 45:23 RWA Foundation Updates 46:52 Token Debrief 58:18 Companies of The Week  

In this wide-ranging conversation, Raoul speaks with Paxos Co-Founder and CEO, Charles Cascarilla, about the evolution of crypto infrastructure, from the early days of Bitcoin mining and blockchain innovation to today's stablecoin landscape. Cascarilla outlines how Paxos is building the neutral, secure infrastructure needed to modernize the global financial system. Recorded on June 25, 2025.

TODAY'S HEADLINESKraken, Robinhood, and Paxos have launched USDG in Europe, unlocking dollar‑pegged crypto for 450 M users under new MiCA rules! Can this shake up USDT & USDC dominance?Stablecoin issuer Circle Looks to Become a Bank, as it applies for National Trust LicenseI teased this yesterday, but its official: Arbitrum will serve as Robinhood's onchain railsTrump's 'Big Beautiful Bill' May Get a last minute Bitcoin Tax ExemptionFinally, Ethereum community plans onchain ‘time capsule' to mark 10th anniversary of network's genesis blockFriends of the ShowC3The C3 team has more than 20 years of experience in journalism, including leading the editorial and content side of a major Web3 news publication. They are also experienced AI and Web3 PR professionals, regularly placing content in leading web3 and AI publications. C3's members previously co-founded the PR department at SCRIB3, and have experience with clients such as EigenLayer, VanEck, Monad, SKALE Network, LEVR Bet, Symmio, Camp Network, Evmos, Avail, Moonbeam, and others.WHERE TO FIND DCNdailycryptonews.nethttps://twitter.com/DCNDailyCryptoEMAIL or FOLLOW the HostsQuileEmail: kyle@dailycryptonews.net***NOT FINANCIAL, LEGAL, OR TAX ADVICE! JUST OPINION! WE ARE NOT EXPERTS! WE DO NOT GUARANTEE A PARTICULAR OUTCOME. WE HAVE NO INSIDE KNOWLEDGE! YOU NEED TO DO YOUR OWN RESEARCH AND MAKE YOUR OWN DECISIONS! THIS IS JUST EDUCATION & ENTERTAINMENT! Hosted on Acast. See acast.com/privacy for more information.

Tune in to this episode of the Security Token Show where this week Herwig Konings and Kyle Sonlin cover the industry leading headlines and market movements, including Grove's $1B allocation, Digital Asset's $135M raise, tokenized stocks and more RWA news!   Company of the Week - Herwig: Dinari Company of the Week - Kyle: Davis Commodities   Companies covered include Grove, Janus Henderson, Anemoy, Centrifuge, Republic, SpaceX, Dinari, Upexi, Superstate, Digital Asset, Canton Network, DRW, Tradeweb, FalconX, Lynq Network, Wormhole, Ripple, Nasdaq, QCP, Davis Commodities, Plume, Blocksense, Jarsy, Baillie Gifford, Archax, BlackRock, Ethena, Securitize, KfW, SDX, MTCM Securitization Architects, Tokeny, Matrixdock, SPiCE VC, Clearpool, Fiserv, Solana, Mastercard, Paxos, PayPal, Particula, Agora, and EFAMA ==== TokenizeThis 2025 Conference Review: https://docsend.com/v/k8bn7/tt25  STM Predicts $30-50T in RWAs by 2030: https://docsend.com/view/7jx2nsjq6dsun2b9    More STM.co Reports: https://reports.stm.co/  Join the RWA Foundation and Read the Whitepaper: RWAF.xyz  Learn More About WALLY DAO: WallyDAO.xyz  ==== ⏰ TABLE OF CONTENTS ⏰ 0:00 Introduction 0:16 Welcome 1:56 Market Movements 18:00 RWA Foundation Updates 20:18 Token Debrief 40:14 Companies of The Week  

Recorded live from the Global Dollar Network event, Network Effect in New York City please enjoy this conversation with Paxos' Head of Product Ronak Daya and CoinDesk Indices' Andy BaehrSee Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

For episode 523, Brandon Zemp is joined by Tarun Gupta, Founder and CEO of Coinshift, a treasury management platform for DAOs and web3 organizations, which manages over $1B in DAO treasuries for teams like Aave, Gitcoin, and UMA. Coinshift empowers both retail users and institutions to unlock the full potential of their assets and treasury by combining capital growth, payments, and accounting software in a single platform, leveraging products like the Paxos-backed, yield-bearing stablecoin csUSDL and its native token SHIFT. Prior to funding Coinshift, Tarun co-founded Parcel, where he served as COO, and gained recognition for his contributions to enhancing the web3 ecosystem. ⏳ Timestamps: 0:00 | Introduction1:08 | Who is Tarun Gupta?2:57 | What is Coinshift?4:04 | Yield-bearing Stablecoins8:00 | Stability of csUSDL11:39 | Institutional interest in Yield-bearing Stablecoins & Assets13:45 | How to buy csUSDL14:46 | Stablecoins for other fiat currencies16:25 | Coinshift services for DAOs & Web3 startups17:14 | Coinshift roadmap18:18 | Coinshift website & socials

Chad Cascarilla is the CEO & Co-Founder at Paxos, a regulated blockchain infrastructure provider and stablecoin issuer. Visit Paxos.com to learn more. Chad is also one of the earliest investors in bitcoin. In this conversation we talk about the early days of bitcoin, the rise of ethereum and other blockchains, the importance of stablecoins, tokenization, helping onboard enterprises, where we are going, and the mission of Paxis. ======================BitcoinIRA: Buy, sell, and swap 75+ cryptocurrencies in your retirement account. Take 3 minutes to open your account & get connected to a team of IRA specialists that will guide you through every step of the process. Go tohttps://bitcoinira.com/pomp/ to earn up to $500 in rewards.======================Xapo Bank, the world's first fully licensed Bitcoin-enabled bank, offers military-grade security with an unmatched blend of physical and digital security, as well as pioneering regulatory oversight, so your funds are always protected. Beyond secure storage, they enable you to grow and use your Bitcoin. Earn daily interest in Bitcoin, spend with zero FX fees using a global card, and make instant payments via the Lightning Network for unrivalled access and convenience. Visithttps://www.xapobank.com/pomp to join.======================Meanwhile is the world's first licensed and regulated life insurance company built for the Bitcoin economy. Protect your loved ones with sound money built to manage life's uncertainty and a broken financial system. Their BTC-denominated Whole Life Insurance policies allow HODLers to pass more BTC on to their loved ones and a tax-advantaged way to access BTC for liquidity during their lifetime. Visit their website athttps://meanwhile.bm/ to join the waitlist for a policy and to learn more.======================