Podcasts about GDM

Last 4 days before regular tickets sell out at AI Engineer World's Fair - this is the single biggest gathering of AI Engineers, Founders, Leaders, and Researchers in the world. Attendees get >$5000 worth of sponsor credits and talk tracks are looking FANTASTIC. Join us!The AI scaling debate always focuses on the question of “how do we get more GPUs?” but the better question may be: how do we make the most of ones we already have.The fact that a frontier lab like xAI could be running at sub-10% MFU (Model FLOPs Utilization) is just a hint at what the real problem may be.For context, older frontier-scale training runs were already much higher than 10%. GPT-3 was around 21% MFU. Gopher was around 32%. Megatron-Turing NLG was around 30%. PaLM reached around 46%. And our guest Anjney says best-in-class MFU today is closer to 60–70%.It's not necessarily that xAI is uniquely incompetent (it's clear they have talented folks) but rather the priorities may be flipped in the GPU arms race.While GPU access is a bottleneck, simply increasing CapEx won't automatically translate to better models as frontier AI is increasingly a systems problem: scheduling, utilization, networking, kernels, frameworks, data pipelines, parallelism, cluster reliability, and the thousand small decisions that determine whether your theoretical FLOPs become real training progress.From building Discord's developer platform and backing frontier AI companies like Anthropic, Mistral, Black Forest Labs, and Periodic Labs to now building AMP's independent compute grid, Anjney Midha has spent years close to the real bottlenecks of AI scaling. In this episode, Anjney joins swyx at Periodic Labs to unpack why the AI race is not just about buying more GPUs, why 95% utilization would have been considered an outage at Google, and why the next era of AI infrastructure has to be more aligned, more efficient, and more responsible.We go deep on AMP's vision for a compute grid that makes FLOPs flow like megawatts, the difference between full-stack AI labs and horizontal pooling, why AI data centers need community buy-in, and how compute markets could evolve into something closer to an independent system operator. Anjney also explains why DeepMind's unpublished research points to a market failure, why end-of-life prediction remains one of the most important AI applications he has thought about for fourteen years, and why “output maxing” may become a new discipline for frontier systems.We also discuss Anthropic's culture, why “luck favors the prepared mind” in coding models, how Claude cracked coding, why too much capital too early can make AI labs fragile, what Periodic Labs is trying to do with science and superconductors, why great researchers can become great CEOs, and why Silicon Valley is both deeply missionary and deeply mercenary.We discuss:* Why 95% utilization was considered an outage at Google* Why AI infrastructure waste compounds at frontier-lab scale* Why “move fast and break things” does not work for AI data centers* How data center backlash, power grids, and community incentives shape AI scaling* AMP's vision for making FLOPs flow like megawatts* Why compute needs an independent system operator* How interruptible demand and dynamic prioritization worked inside Google* Why DeepMind research hoarding creates negative externalities* AMP's 1.2GW base-load ambition and the need for 6GW of spike capacity* Why end-of-life prediction could become one of AI's most important healthcare applications* Frontier Systems, output maxing, and full-stack alignment* Why APIs and abstraction layers become lossy as organizations scale* Superconductors, standards, and the dream of lossless systems* SF Compute, open protocols, and the future of compute marketplaces* Why non-NVIDIA chips can still benefit from NVIDIA's reference architecture* Trust boundaries and why chip startups need visibility into future model architectures* Why VCs often underestimate researchers as CEOs* Scientists as star athletes of the mind* Why great CEOs need to be confrontational up and down the stack* Why leading the frontier matters more than “winning”* How Anthropic cracked coding* Why culture is fragile, not a permanent moat* Why hardship was a feature, not a bug, for Anthropic* Why Anthropic's P0 was coding from day one* Periodic Labs, physics as the constraint, and technical reality* Silicon Valley mercenaries, missionary teams, and what happens after a breakthroughAnjney Midha* LinkedIn: https://www.linkedin.com/in/anjney* X: https://x.com/AnjneyMidhaAMP PBC* Website: https://amppublic.com/* X: https://x.com/amppublicTimestamps00:00:00 Introduction00:00:09 Why AI Compute Is Being Wasted00:03:17 Responsible Infrastructure and Data Center Backlash00:06:07 AMP Grid: Making FLOPs Flow Like Megawatts00:12:41 Foundry, Frontier Labs, and Research Hoarding00:14:42 Gigawatt-Scale Compute and End-of-Life Prediction00:24:08 Frontier Systems, Output Maxing, and Alignment00:27:38 Compute Markets, SF Compute, and Non-NVIDIA Chips00:32:57 Trust Boundaries, Co-Design, and Researcher CEOs00:38:17 AI Coachella and First-Principles Thinking00:42:43 Leading vs Winning in Frontier AI00:45:54 How Anthropic Cracked Coding00:48:25 Culture, Hardship, and Anthropic's P000:54:03 Periodic Labs, Physics, and Silicon Valley Mercenaries00:56:26 Rishi Valley, Singapore, and Money as a Measure00:58:47 Closing ThoughtsTranscriptIntroduction: Anjney Midha, AMP, and Compute WasteSwyx [00:00:00]: We're in Periodic Labs with Anjney Midha, CEO, founder of AMP. Welcome.Compute Utilization: Node Allocation, MFU, and AlignmentAnjney [00:00:09]: Thanks for having me. At Google, there are two types of utilization usually, right? That you're measuring in these clusters. One is node allocation, and then the other's MFU. Node utilization is usually like what percentage of cards in the data center are just, used, and that, if it's not at, 95%-Swyx [00:00:29]: There is no excuseAnjney [00:00:29]: There's no excuse, right? I think 95% at Google, which is where my co-founder, Seb, came from, he built the Borg, PBorg/GQM scheduler at Google, and there I think 95% was considered an outage, so 96% node utilization is, should be standard. And most single-tenant clusters are not running at that. So that's one. And then MFU should be, I would say the best in class today is somewhere between 60 and 70%. I think this is a leadership question, right? Fundamentally it's an alignment question, which is are the people who are funding the cluster and then deploying the cluster actually aligned? And sometimes theoretically they are, but in practice the number of people in the chain, the supply chain between, the capital and all the way to whoever's managing the cluster and then whoever's measuring what the output is, are just so many, degrees of separation away that, the, The Have you ever heard the radian metaphor, which is at the beginning of an arc, if you have two arcs that are two lines that are just off by a few degrees, that-Swyx [00:01:33]: It spreads outAnjney [00:01:34]: It spreads out, right? Or at scale. And I think what's happening is a lot of cluster implementations and infrastructure, a lot of frontier labs and other teams, that's what's happening, is they're, they initialize the plan, which is kind of like North Star with a team that wants to do good, but then they're, required to scale so fast instead of iteratively that the wastage just compounds really fast at scale. And so I think we know the answer, which is just do iterative bring ups. If you spend time with people who've been in the semiconductor industry or the DSN industry for a long time, this is not new, and I don't think AI should be an excuse. Sure. Something What is new? Okay. We have a lot of new capabilities, but that doesn't mean just abandon common sense. Common sense should always be in fashion. ? AI scaling doesn't change the in fact, if anything, AI scaling should be putting a premium on the value of common sense and infrastructure because the margin of error now is so much lower and the costs of wastage are so much higher. And the cost of wastage, by the way, is not just economic. I'm, obviously I'm, I'm an investor, or I'm an investor by background. Over the last few years now we're running an AI infrastructure business called, AMP. And I think that it's okay to say this time is different on the capabilities front. We are genuinely getting capabilities at, of the, of a kind we haven't had before. That doesn't give you an excuse to say this time is different for everything, especially infrastructure. So look, I love the hacker mindset and the hustler mindset. Now, that's great for the startup mindset, but you remember this moment where Zuck went from saying, “Move fast, break things” to, move-Responsible Infrastructure and Data Center BacklashSwyx [00:03:10]: Fast and stable infrastructureAnjney [00:03:11]: Move fast with stable infrastructure. I think now we need to move fast with, responsible infrastructure. People are going to ask where the impact is. There was a really In our class yesterday, Scott Nolan, who's the founder of General Matter, came by at Stanford to speak about energy bottlenecks. And he had a phenomenal idea. He said, “if you look at the marginal unit economics of compute per hour,” he goes, “let's call it, $4 an hour. If you're having to bring up a new data center in a new community, why not just say we're going to charge 4.50 an hour, and that marginal impact or that marginal increase, we just literally take that and give it to the local community as cash?” I can tell you as a customer of that compute, I would love that. I'd be happy to pay an additional 50 cents per hour at scale.Swyx [00:03:57]: Wow. Yeah.Anjney [00:03:58]: Because if that means the public benefit is so clear to the communities that the data centers are coming up in, I'm going to feel like that compute is much more reliable. Up to 20% of all data centers this year in the US, my understanding is are at risk.Swyx [00:04:13]: Of community backlash?Anjney [00:04:14]: Correct. Of not getting the community support they need to get brought up.Swyx [00:04:19]: Wow. That's a huge number.Anjney [00:04:20]: Yeah. Now, we, I think we should dig into what that number is. I think it's a little bit of overstated. These things can get over-reported, but it-Swyx [00:04:27]: They don't just care about jobs. They care about all the other stuff around it, right? They care about power grid, they care about environments-Anjney [00:04:33]: Power grid, permitting, and so on. And imagine I think if you said there's a new AI deal. If we're bringing up a data center in your community, we're actually going to reduce the cost of your electricity bill. Okay, now we're talking. Right? The community's going, “Okay. Now this is a deal. I feel like a partner in this.” Right now that's not happening. There will be audits, there will be investigations, and when the, when the regulators come, I don't know when it's going to be, the folks who are moving fast and breaking things in the name of AI progress better be prepared. That's certainly not how we're procuring compute. Or we're, we're trying as much as we can to work with partners who have long-term track records. Many of whom, by the way, are not, AI providers. I think this whole idea of neoclouds being somehow this new category is a lot of marketing speak. There are really good, reliable, trusted data center providers in America who've been around 20 plus years. I love those folks. They know how to Sure. Are they sponsoring happy hours at NeurIPS? No. Are they legibly listed in Build? No. Are they hanging out in my, in, situational awareness parties? No. But they're adults. I trust them.Swyx [00:05:44]: They can run LAN. They can run power.Anjney [00:05:45]: They can run LAN, power, and shell. They have credit histories. We sit down, we have a conversations. Many of them live in Silicon Valley. They've, they've had to deal with the boom and bust cycles of the internet, and I love those folks. They are stable infrastructure partners and thinkers. And I think there's a lot of short-term thinking going on in the compute layer, and it's going to catch up to us. It's not going to be good.AMP Grid: Making FLOPs Flow Like MegawattsSwyx [00:06:07]: You talk about aligning incentives, and, I would think that aligning incentives means you have the full stack in one company, which is xAI and OpenAI, right? So you as a standalone infrastructure layer, why are you somehow more aligned to your portfolio companies than people who just own the whole thing?Anjney [00:06:28]: In systems design, right, there's, there's two regimes of, architecture, right? You have integration, and then you have pooling and utilization, right? So the Or rather, the way to increase utilization often is you can do systems integration where you collapse a lot of process into one node, or you can pull out a process from a node and share that amongst various That resource amongst several different nodes. And so we see the AMP grid, which is, the, what, the system we're building here, which is basically a compute grid. We're trying to do for compute what the electric grid-Swyx [00:07:02]: PowerAnjney [00:07:02]: Yeah, what the power grid did for electricity. It-- this is a pooling and utilization layer across clouds, And so we're actually the opposite of a full stack integration like approach.Swyx [00:07:12]: Super horizontal.Anjney [00:07:13]: Where it's much more horizontal and it's, it's multi-cloud, it's multi-silicon. The goal is to try to make FLOPs flow like megawatts, and that is very hard to do today for many reasons. There's stranded pools of compute all over the place and there's no fungibility. And so right now we do it at the level of scheduling, and we often do it at the economic layer. But as we start to announce what we're working on, it's extraordinary like how many folks are coming out of the woodworks and saying, “Hey, I'm actually working on a way to make compute fungible at this part of the stack and that part of the stack.” And as a grid, we'd like all of these folks to participate on the grid. There's, people often ask me, “Andra, are you a new cloud?” And I go, “No, actually neoclouds are suppliers.” sometimes they'll ask, “Are you a venture capital firm?” I go, “No, actually they are, they are demand like sort of off-takers of the grid.” We see ourselves as what's called an independent system operator. So if you study the history of the electric grid, once it became legible to a lot of factories and industrial sort of participants that, hey, actually it turns out pooling is a good idea. We should pool our generators instead of all having a generator running at half capacity in our backyard. There was a need for an independent entity who could coordinate all these parties. Transmission line, power generation, facilities, transmission lines, factories, and that neutral coordination mechanism is very critical. In order-- If you study like the history of grids, the most enduring ones were those that never owned their own assets. They were ones that had, or often started with long-term anchors who are uncorrelated sources of demand, a steel factory, a shoe mill or whatever in a particular town who weren't competitive, where the steel factory want to spike up at night, the shoe mill wanted to spike up during the day. So then you pool and you share, right? So each of you is guaranteed some base load, but then you kind of schedule your spikes to drive a peak utilization across the town. The gold standard, so to speak, historically, has been these utility companies like PJM Interconnect in the northeast of America, where they, over many years became this what's called an ISO, an independent system operator of the grid. So that's how we see ourselves. Economically, that's what we are. From a technical perspective, we started at the scheduling layer because Seb and Mihai, who, run engineering here, built that at-Swyx [00:09:28]: Did your schedulingAnjney [00:09:28]: They did that at Google. And, -Swyx [00:09:32]: And you have infra shops from Discord as well.Anjney [00:09:35]: I have some.Swyx [00:09:35]: I don't know, I don't know if Discord is like the primary identity, but what-whatever, I'm just kind of-Anjney [00:09:39]: No, D-Discord was-Swyx [00:09:40]: Choosing a well-known name.Anjney [00:09:42]: Well, I So I was running the developer platform there. The internal infrastructure I was not responsible for. That was actually a guy by the name of Mark Smith, who was extraordinary. And yes, Discord did pool So Discord is actually a counter example. I had the chance to learn a lot about fully, full stack infra there because-Swyx [00:09:56]: It's the same thing, yeahAnjney [00:09:57]: It's the, it's the other architecture which is, Discord built its own WebRTC vo-voice and video infra. So like Discord did not use-Swyx [00:10:08]: For the calls, yeah.Anjney [00:10:09]: Yeah, did not For communication, Discord did not use third party infra. It was all built in-house. And then the way you maximize utilization was you pool demand from the world's 200 million plus monthly active gamers, right? And so that's, that's how those stacks were constructed. Again, in systems design, the two concepts that keep coming up over and over again are abstraction and composition, right? And-Swyx [00:10:31]: Bundling and unbundlingAnjney [00:10:33]: Bundling and unbundling, abstraction, composition, like verticalization and-Swyx [00:10:36]: HorizontalAnjney [00:10:36]: Horizontalization. So in that sense, AMP is an independent system operator of the grid. We pool demand, we pool supply from a number of partners we trust At about 1.3 gigawatt scale over four years. And then we pool demand from some of the world's best, research labs and so on. We're sitting at one, periodic labs who need extraordinary long-term demand. And the idea is that, each of them is guaranteed base load on the grid, but they can spike up and down flexibly on, for compute, with much shorter timelines as needed. That was roughly the design of the program I came up with at a16z called Oxygen. The same-- That was the same design of the GQM, BorgX, Borg GQM implementation at Google that Mihai and Seb had built. Which was that how do you allow, teams inside of Google, on the internal infrastructure to be guaranteed capacity, for their base workloads? But when they need to spike up on research, how could they ensure that was sufficiently there? And of course, the big innovation that was not discovered, but kind of implemented in the space, this infra space maybe three, four years ago at Google was the idea of interruptible demand, right? Where you just queue up a bunch of jobs and through this like sort of credit system, there can be a bidding mechanism.Swyx [00:11:53]: Like priorities.Anjney [00:11:54]: It's a dynamic prioritization Basically. And jobs can get interrupted based on somebody else who's saying, “what? I have 10 tokens, 10 credits I want to spend on this job.” Another like team lead, research lead is “Genie 3 or whatever is only worth five, credits, and NanoBanana2 is worth 10 credits,” and so the NanoBanana job gets priority. That's a, that's a made up example.Swyx [00:12:15]: It's very real. Brain Marketplace was real. And, we've, we've covered this on the pod with David Luan, who was-Anjney [00:12:20]: Oh, great. OkaySwyx [00:12:20]: Was there. And the criticism is that, well, actually sometimes you need central command to go all in on a thing. And actually sometimes capitalism via credits doesn't work. Not, this is not a criticism of AMP. I'm just saying, this is a thing that has been tried, internally within Google, and it led to Google missing GPT.Foundry, Frontier Labs, and Research HoardingAnjney [00:12:41]: Like, we structured ourself essentially very similarly to Google. We are structured as a holdings company. So, Alphabet holdings is Alphabet holdings, and then they've got these subsidiaries called Google and-Swyx [00:12:51]: Other betsAnjney [00:12:52]: Other bets and so on. We've got, AMP holdings, and we've got our infrastructure business, and then we've got a capital business called Foundry that incubates new frontier AI labs or invests in them as venture capital, like Periodic. We put a few hundred million dollars into Anthropic from our fund earlier this year. So wherever we feel like teams are making progress, especially researchers and so on who've pushed the frontier inside of existing labs like DeepMind, I find, there comes a point where they feel misaligned with the dictatorship of Alphabet holdings. And at that point, sometimes the dictatorship doesn't want them anymore. And they're “Thank you. You've done your job here. You've kind of helped us through the zero to one phase, and for whatever reason, we're going to deprioritize your amazing, omni model or whatever it is, and instead we're going to prioritize coding.” And, I think that's a tragedy, but I get it. They're Sergey and team are running their own business there. But that doesn't mean we the rest of us should sit around waiting for that progress to get unlocked for the rest of the world and humanity. If you think about how much extraordinary research has happened inside of DeepMind over the last 10 years, I, Demis and Sergey and those guys did such a great job. But at the end of the day, so much of that has never seen the light of day?Swyx [00:14:00]: Or they're like papers only, but they never actually shipped it to production or-Anjney [00:14:03]: What's worse is the paper is actually not even being published anymore ‘cause there's a six-month embargo inside of DeepMind, right? We've heard about this where a paper comes out, and then I think there's a six-month embargo window where if anybody on the business team says, “This could be interesting” It's embargoed for life.Swyx [00:14:18]: Exactly. So the stuff that gets published is the stuff that's not good enough.Anjney [00:14:21]: There's an adverse selection problem, basically. Yeah. At this point-Swyx [00:14:25]: It's, it's a common complaint at NeurIPS, by the way, that's “Well, why would I look at the papers that are the trash of GDM?”Anjney [00:14:31]: Again, I think it's a tragedy. I get it. They're running their business, but the rest of the I think there's negative externalities of research being hoarded, and so that'there's a market failure. And somebody needs to unlock that research, and we can't do it on our own. We only have 1.2 gigawatts of compute. That's nothing. That's about $40 billion of cloud spend. We're going to need a lot-Gigawatt-Scale Compute and End-of-Life PredictionSwyx [00:14:51]: By the way, is that's a new number. I haven't, haven't come across that gigawatt number. That's huge.Anjney [00:14:56]: Yeah. And to be clear, we haven't secured all of it. That's how much demand we have started to secure. I think publicly we haven't actually confirmed how much we have for this year. In order-Swyx [00:15:04]: Where do you want to get to?Anjney [00:15:06]: I think the steady state would be that we have a base load pool Of 1.2 gigawatts at all times Of base load capacity. For spike capacity, right now my estimate is we need roughly six gigawatts over the next four years for all our teams to feel like they were able to keep moving the frontier, whatever they're working on, whether it's, like superconductor discovery over here. There's a new investment we're working on right now, which is in the end of life prediction space in healthcare. It's extraordinary how much you can, you can give this was actually my graduate school work. I went to grad school for bioinformatics at Stanford Med. And I know we-Swyx [00:15:40]: Econ, MCS, bio.Anjney [00:15:41]: So my-- I was this really weird cat where, I was never satisfied with my major options. So at one point I was an econ major, then I was a CS major, then I was a MCS major called mathematical computational science, and they decided they were going to end that major. So I took all that coursework, and I applied it to grad school, my graduate degree in bioinformatics, which was the master's program, and then I thought I was going to do a PhD. I never ended up doing it. I dropped out and went to work at Kleiner. But I was lucky enough to apprentice with this professor at, Stanford Med. His name is Nigam Shah, and he was working on end of life prediction. Stanford is one of the only research facilities in America that has a longitudinal patient data set that's larger at scale. I think it's at least 12 million patient lives. The only larger data set is at the VA, the Veterans Affairs, of America. And to do research, like do any deep learning and so on that data set, it was called the STRIDE data set at that time, you had to be a Stanford Med School affiliate, which is why I went and enrolled in the bioinformatics department. End of deep learning was early. Nigam Shah had the visibility-- the vision to see that, you could do end of life prediction to help palliative care. In America, the, over 30% of all Medicare, Medicaid spend, at least at that time, was spent on end of life care. And what's we grew up in Asia, so we all-- Yeah, at least I won't speak for you, but I have A very different relationship with death than I find folks who grew up in America do. In America, spiritually and culturally, especially in Western societies where Christianity, the Christian tradition sort of frames death as this terminal point, there's often a judgment day and so on. The way we view death is with a finality. In Indian culture, in Hindu culture, death is one-Swyx [00:17:35]: Also, he's Buddhist as well.Anjney [00:17:36]: You're Buddhist, yeah. So it's one, it's one step in a journey of many lives, right? And so, I grew up in this city called Chennai in the south of India, and when people die, you dance on the street. There's like a procession where your body is carried to be cremated and your family, like celebrates and there's drums and so on. It's this huge thing. And, It's because the idea is that you're going to be reincarnated. You've been liberated from the responsibilities of this life, and now you're onto your next. It's a new It's like going off to a new college or whatever, right? And so it was so alien to me when I got here as an undergrad- That the medical system works backwards from that assumption that we have to view death as this terminal thing and delay it, postpone it's a bad thing. And so at the time, clinical decision support in the United States was this very primitive field. Even to this day, physicians in the United States often will tell you when you have a terminal disease, this is your, we've diagnosed you, which is great. Our ability to diagnose you is extraordinary. You have somewhere between six months to six years to live. What do you do with that information? The error bars are so high that then you In times of uncertainty, we default to culture, and when the culture is let's-- this is a bad thing, I've got to prolong my life, then you start doing things like And just to, just sort of from a systems perspective, what's going on there is Physicians often feel like they need to provide such high error bars because there's always some uncertainty in end of life diagnosis, and if you provide the wrong Diagnosis or recommendation to your patient, you can be sued for medical malpractice. And then your license can be taken away. It can be catastrophic for your career. In contrast, if in countries where that's not the case, what you often observe is that patients, physicians are quite prescriptive with their recommendation. They say, “Hey, this is your condition. The literature says that you probably have this much time on Earth left. My expert opinion is that you are an outlier or whatever.” And they try to be more prescriptive, and that empowers a patient, right? ‘Cause then a patient can say, “I trust my doctor. They said on average, I have six months to live, but if I do these things, I may have a shot because of my particular predispositions or my genetic history or whatever.” And that empowers you to go about your life in a actually more scientific way than leaning on religion, culture, spirituality, and so on. In contrast, here, because of that medical malpractice sort of thing looming over your head, a physician never gives you a clear recommendation. So instead you say, “Okay, Doc, well, let's try it all.” And then you start a whole regime of drugs and therapies, and then you often spend weeks and weeks in the hospital, and that deteriorates your quality of life. And when that deteriorates your quality of life, you instead of spending your last few days doing the things you love with your family, you're spending it on a hospital bed. And that ends up being thirty percent of Medicare and Medicaid. So it's worse for the patients. The doctors feel terrible. The American taxpayer is paying a huge amount of money. And so this is why Nigam Shah, who was this professor at Stanford, said, “Anjney, if there's “ I kind of sat down with him. I was this young, I'd, I was twenty-one, and I was “I want to work on a big problem.” He's “The big problem is end of life care.” And so we tried to do deep learning to say, to-- So we started trying to run deep learning on these tried patient data sets to say, “Could you have an AI system make a recommendation that is orders of magnitude more precise about how much time you have left once you've been diagnosed with a terminal condition than a human?” And then if we can get that precision to be high enough, then you can empower the patient. And it turns out the tech works. Like it's-- Once you get the data set, like RL works. Honestly, even regression models work. You don't need to get that fancy. At the time, we were just trying, doing like very simple neural nets.Swyx [00:21:54]: Simple solutions, yeah.Anjney [00:21:54]: Today, what we can do with RL is extraordinary. The problem remains then and now is regulatory, because you actually can't shift the burden of the wrong clinical diagnoses from the physician to the AI system. And so at that time, I got quite disillusioned ten years ago for, twelve years ago where, ‘cause I felt I just didn't have the resources to influence regulation. Today, I'm very lucky. I'm in a different place. I've, I'm a lot older, and so I've been spending a lot of time on my next incubation, which is how can we unlock the, patient empowerment by training AI models to do end of life prediction much, with much more precision and ac-Swyx [00:22:37]: Oh, wow. You're still focused on this the whole time.Anjney [00:22:40]: The-- I haven't been able to get, this out of my mind a single day for the last fourteen years. This is the hill I want, I would like to die on. There's two, I would say. What? I actually, I'd prefer not to die.Swyx [00:22:51]: Yeah, exactly.Anjney [00:22:52]: But I think two bipartisan issues, I think two issues that should be bipartisan in America are how do we empower patients to make the right clinical decisions at the end of their life, such that we're reducing the taxpayer burden with science? It's just good old science, and AI can help here. And the second is, net positive data centers, ‘cause I think that's the biggest critical bottleneck on training and good enough AI models to help people at the end of their life. So there's sort of two sides of the, of the same scaling bottleneck curve, but those two, we formed AMP as a public benefit corporation. My wife and I, who you've met, you've met Viv. Her passion is education. Her family is a long line of educators and so on, and, of physicists. And so this class is my attempt to stop being the black sheep of the family and be a, an educator. But if I'm not educating, the thing I would be doing is working, on these two problems, whether on the political spectrum or as a researcher back at, in some lab. And my hope is if anyone's listening to this podcast, if they're passionate about either of those two topics, I'd love to hear from them. We'll, we'll we can share the contact in the show notes, but, we're looking for people to join both of those missions on the, on the political side as well as on the medical side, on the research side.Frontier Systems, Output Maxing, and AlignmentSwyx [00:24:08]: You said, this is a discipline that you want to form. You call it's called variously called Frontier System. It's variously called One Person Frontier Lab. What is the ideal name or shape of this? Like the, what is the mission?Anjney [00:24:24]: Of the class?Swyx [00:24:26]: Of the discipline that you're, exploring, right? I The class is called Frontier Systems. But like for me, maybe one phrase is you're, you're just anti-waste, right? Which is wasting GPUs, wasting in human and Medicare. But is there, is there a broader theme that I'm, that maybe you can encapsulate more succinctly?Anjney [00:24:45]: Yeah. The, from an engineering perspective, it's very simple. It's output maxing. It's the, it's the department of output maxing.Swyx [00:24:51]: Making the most of what we have.Anjney [00:24:52]: Exactly. I'm a huge believer in optimal outcomes. I think both in America and other countries, we are losing our appreciation for nuance, and this is the thing of And AI is the same case, right? Oh, the bitter lesson holds. Okay, fine. But that doesn't mean you just like throw 500 GB300, 500,000 GB300s at your suboptimal model scaling and you waste a bunch of compute. It also doesn't mean that, the most optimal is to have like 50 different architectures where there isn't enough standardization. One of the reasons Anthropic has had extraordinary sort of velocity is ‘cause they picked the transform architecture and said, “This is simple. Let's double down on it,” right? And now luckily there's enough investment going to the space that we can afford other architectures, but at the time, investment was just too fragmented into other architectures, so that arguably unlocked scaling. So I think there's a philosophy. I think we all owe it to ourselves to do output maxing with a new capability called AI on a global level. I think if I was starting a new department at Stanford, depending on how fuzzy or technical I wanted to be, I'd probably call it the Department of Alignment. Like-Swyx [00:25:59]: It's an overloaded termAnjney [00:26:01]: But it is, But alignment really Is a hard problem. And I think when you unlock it, full stack alignment is super hard in any organization and in any system. Like in a, in a venture capital firm, if you can have full stack alignment between your limited partners and your, the founders who are creating the value and ultimately the public that owns the IPO stock, that is a gift that keeps giving. And when you study the history of these systems, when they start off, they usually start out small scale where the feedback loop is actually so tight that there's alignment. And then the more you try to scale, the more division of labor happens, the more specialization happens, and at each step you add abstractions. And wherever there's an API interface, there's like loss. There's communication loss. And so I think a really cool thing would be for us to figure out is there a way for us to have our cake and eat it too as an engineering discipline? Is there a way to actually scale up and scale out Without losing any alignment, without lossy transmission?Swyx [00:27:01]: You mean standards?Anjney [00:27:02]: So standards is one way. The other way is you just have net new capabilities. So like what we're trying to do here is discover new superconductors. A room temperature superconductor would be a lossless transmission mechanism for energy. We would have flying cars. We are right within a few years of having a new room temperature superconductor. So I think those are the two. You either have to standardize On protocols or API specs that allow lossless communication, or you can come up with a whole new capability that unlocks so much abundance, the standardization doesn't matter ‘cause you just unlock net new capacity. This, the, so this is what I spend my days thinking about these days.Compute Markets, SF Compute, and Non-NVIDIA ChipsSwyx [00:27:38]: No, I think every infra person at, who wants scale and wants to output max does eventually end up thinking about this. We don't have time to go into it, but we have done an episode with SF Compute-Anjney [00:27:50]: Oh, coolSwyx [00:27:50]: That is trying to standardize The futures contract for compute. I don't, I don't know how that's going by the way, but like at some point this will be public.Anjney [00:27:57]: Oh, I think Evan is awesome and SF Compute is the kind of effort that I hope we can accelerate because what often happens is these exchanges are very hard to get, they, it's hard to bootstrap them, right? Because they often require-- There's many inefficiencies between parties. There's trust boundary inefficiencies in infrastructure because you don't trust, one part of the stack doesn't trust another part of the stack to give them visibility. There's capital markets inefficiencies, there's operational efficiencies. So if you can inject like a single shock to the system of a ton of compute demand or supply, then you can accelerate, these new flywheels. And so my hope is one day, or soon, if SF Compute needs extra like has excess capacity, they just hook it up to the grid and they get flooded with demand from us. And on the other side, if they have a ton of demand but they don't have supply, they just again hook up to the grid and it's a two-way protocol where they can just hook up to our capacity. And I don't think we're too far from that. Today our working implementation of it is mostly through a group of labs, universities, and a few sort of trusted parties who are, who all feel like they're in alignment to borrow an over sort of used word. But our hope is to just have it be an open protocol that anyone can hook up to on-Swyx [00:29:20]: Hook up for demand or hook up for supply? In primarily demand, it sounds like. Like you-Anjney [00:29:25]: No, bothSwyx [00:29:26]: You would want to offer demand.Anjney [00:29:27]: Both. Yeah. Unfortunately, what's happened in the last six weeks is, we thought we'd have a bunch of excess capacity by the end of this year. It's all gone.Swyx [00:29:37]: It's exploding.Anjney [00:29:38]: It, yeah. It's all gone. And so I have, my text messages are full of friends, we know many of these people, these are founders who've raised billions of dollars in San Francisco going, “Oh, any chance you have like 50 nodes in the next few weeks?”Swyx [00:29:51]: What is the scope for, non-Nvidia, right? You have Lisa Su coming and, Rainer Pope as well. And so There is a lot of demand for, more performance Alternative architectures and all that. At the same time, this hurts your standardization.Anjney [00:30:11]: I don't think so. So actually Rainer's a great example, right? Rainer is a CEO and founder of, MatX. I actually had him by for office hours in the class earlier today, and there was an insight he brought up that I hadn't considered before, which is when they decided to pick the standard For their data center, they picked the NVIDIA reference architecture. So the MatX chips Just plug in to any site that has an NVIDIA bring up planned. And, the-Swyx [00:30:42]: It's just software then. It's, it's not the-Anjney [00:30:44]: A-Swyx [00:30:44]: Hardware.Anjney [00:30:46]: Well, from an input and IO perspective It's the same footprint as an NVIDIA rack.Swyx [00:30:52]: That makes sense.Anjney [00:30:53]: Where they have done, innovated a bunch from what I can tell is on systems co-design. Which is where a lot of the gains are to be had. And so he picked He was “Anjney, we, there's just so much work to do when you're building a new chip company.”Swyx [00:31:08]: Can't fight every front.Anjney [00:31:08]: You just can't fight on every front. So my question to him was, “Well, you're working on this new chip. Their tape-out is next year. What, who are you going to partner with to host the chips?” And he said, “Whoever will host them. That's just not, that's not my focus.” And I said, “But how did you “ you decided back to our earlier systems design question, he decided that, he didn't want to be a full, fully integrated chip provider. The bottleneck they're focused on is the logic die, and they, he feels they can crank out a ton of performance gains through co-design there. But then that means you delegate, to our question earlier, it, you he's the data center provider is a different part of the stack, and so then he's dependent on that part of the ecosystem to host his chips to get the performance gains to the customer. So now you have another abstraction, and you might have loss. So I asked him, “How do you prevent loss?” And back to your point, he said, “I just picked the NVIDIA standard ‘cause I didn't want to Like I wanted to piggyback off of an existing protocol.” And that, what's great about NVIDIA is that reference architecture is known.Swyx [00:32:15]: Open.Anjney [00:32:15]: It's open. They've published it. So Jensen's actually enabled someone like Rainer to build a chip company like MatX, and I don't see them as competitive. The compute demand is so high. Like, I don't I think NVIDIA's not able to meet the demands of production, so we just need more chips. And I think it's very smart what MatX has done, which is say, “We're just going to we're not going to innovate on the data center design ‘cause actually, thank you, Jensen, you've done all the hard work. Where we can innovate is somewhere else.” And I think that's, that's very healthy. I think that's how we unblock new bottlenecks. And my view is these, the, chip teams like MatX, who have arrived at the insight that co-design is the way, The primary bottleneck for them is trust boundary. To do co-design well, you need visibility into the next model generation as soon as possible ‘cause it takes two years to tape out. So if by the time I bring my chip to market, your model architecture's changed, I'm host. Now, when he was inside Google, he was sitting next to the Gemini team. He was on Palm or whatever.Trust Boundaries, Co-Design, and Researcher CEOsSwyx [00:33:19]: His co-founder was the, was one, was one of the Palm guys, I think.Anjney [00:33:23]: Yes. Yes, exactly. So when you're inside the trust boundary of Google, then your systems co-design loop is super tight. When you leave as a founder, one of the biggest risks you take is now you're outside the trust boundary. And so what I love doing is helping chip teams who can help us unlock more capacity for the independent ecosystem access to trust. Because when I If I've been, involved with a lab from day one, and I was lucky enough to work with Anthropic, and then I'm on the board of Mistral and helped Black Forest Labs get started. I think at this point I'm on six or seven different teams.Swyx [00:33:57]: Only six? I feel like my mental number was going to be 13, but yeah, it's-Anjney [00:34:02]: No, I go deep with one at a time.Swyx [00:34:04]: You're founding CEO of Arena.Anjney [00:34:07]: Nah, that was an, that was an-Swyx [00:34:08]: Administrative CEOAnjney [00:34:09]: It was an administrative five-month gig where Whalen and Anastasios were graduating from their PhDs, and they didn't need a product team. So I helped recruit the head of engineering product and design. But Anastasios has always been the CEO of that company. I played a pinch-hitting I'm an intern. I was CEO intern For five months. -Swyx [00:34:33]: I interviewed him, and he's he's very well-spoken. I think he's a debate, former debate, champion. But also very quantitative and mathematical, which is-Anjney [00:34:41]: He-Swyx [00:34:41]: Such a unicorn.Anjney [00:34:43]: See, what's amazing about him? If you look at his output, he's an output maxer. By the time he was graduating from his PhD, which he only graduated last year, he had published more work with a citation count than, people twice his age. But at the same time, he'd already started a project called LLM Arena that was being used by millions of people As a side project. And time and time again, what I've realized is venture capitalists suck at seeing human beings as, dynamic agents where-Swyx [00:35:14]: They want to put you in a boxAnjney [00:35:15]: They want to put you in a box.Swyx [00:35:15]: This is your thing.Anjney [00:35:16]: So the first time I got introduced to Anastasios, somebody had told me “Oh, he's amazing, but he's a researcher.” I was “what? What do you mean he's a researcher?” That's what-Swyx [00:35:28]: Like he's not a CEO, not a founder.Anjney [00:35:29]: Not a CEO, exactly. I was “Are you crazy? Do you Have you met Dario?” Dario's a scientist. He's gone from zero to, what will soon be a trillion-dollar company in four years. Being a CEO, nominally speaking, is not that hard. Being a good CEO is hard. Being a great CEO actually requires a level of performance that scientists who have already published at the top of their field have accomplished. It is super hard to be a competitive scientist. To publish in academia over the last 20, 30 years, to make it to the top of your discipline at a place like Berkeley, you are a star athlete. Like, you are an athlete of the mind, and you perform at the highest levels. And to get there, whether you're, Anastasios or Whalen at Berkeley, or you are Robin, who-Swyx [00:36:23]: BFL, yeahAnjney [00:36:24]: With Black Forest, who created Stable Diffusion, or if you're, like Guillaume at Meta, who created Llama before he started Mistral. The amount of human leadership you have to demonstrate to get the resources, like get the trust of the organization, publish it, put it up. I would just fund researchers all day Right? If who have contributed already to the field. If they've, if they've put SOTA out there, they're, they're star athletes already. If they haven't done SOTA Look, they can still be good CEOs, but then I find the failure mode is that they just don't want to be CEOs, they primarily want to publish, and that's okay, too. One of the things we do with the AMP Grid is we donate excess compute. We have two nonprofits, like university labs. We carved out like a couple thousand H100s. But I do think there's extraordinary research being done on university campuses. My father-in-law's a physicist. He's a professor. Extraordinary work in physics, and we need that. But if you want to be a CEO, what you need to be willing To do is be super confrontational, outside of science. Like within the scientific community, some of the best researchers are very confrontational about their convictions, right? This architecture is right. To be a great CEO, you basically have to be willing to be confrontational up and down the stack.Swyx [00:37:41]: To your own team.Anjney [00:37:42]: To your own team-Swyx [00:37:43]: To customersAnjney [00:37:43]: Hiring, recruiting customers. Well, I would say, Yeah, pretty much to everyone Everybody. Of course-Swyx [00:37:50]: I see, I feel a little bit of that in my own work, but yeah, I can't imagine the stakes that Dario has had to go through. It's, it's pretty insane.Anjney [00:37:56]: No, I don't think the stakes are that different From how you're feeling it, right? Stakes are personal scaling vectors, right? The stakes that seem so low to you, like having this podcast where you can talk to somebody and just have a you're an extraordinary communicator, right? Like already in this conversation, you've pulled more out of me than most people, and I've been on 12 podcasts in the last two weeks.AI Coachella and First-Principles ThinkingSwyx [00:38:17]: I think I, we've just seen each other enough that there's some base trust.Anjney [00:38:20]: There's base trust.Swyx [00:38:20]: And I think, and I know that you, that I've done my homework and like I know that trust is a big deal for you, so.Anjney [00:38:27]: I think trust is about consistency, and you and I have seen each other In the community for years, right? Like, I remember the first time we met was at NeurIPS in New Orleans. I don't know if you remember that, luncheon.Swyx [00:38:38]: Oh my God.Anjney [00:38:39]: Reiko had set up this Reiko's amazing, and he set up this luncheon and-Swyx [00:38:43]: Yeah, I was “Who's this Discord guy?” I'm “Okay.” But-Anjney [00:38:45]: No, you weren't-Swyx [00:38:46]: You were just “You made some investments.”Anjney [00:38:47]: You were much less polite. You were “Who's this VC?” You're like-Swyx [00:38:51]: No, I Was I? Oh my God.Anjney [00:38:53]: It was-Swyx [00:38:53]: I'm so sorryAnjney [00:38:53]: It was visible on your face.Swyx [00:38:54]: I'm so sorry. But you weren't, you weren't The introduction was bad. I was I didn't know who you were.Anjney [00:39:00]: The, see, this is the thing about context, right? Like, but then I think I heard your accent. And I was “Are you-”Swyx [00:39:06]: Singapore, yeahAnjney [00:39:06]: “Are you Singaporean?” And you're “Yeah.” And I said, “I went to high school, JC, in Singapore.” And then the ice broke. But This is the there are in the scientific community, sometimes the stakes are very high for people who haven't had the emotional, what is called EQ Coaching and mentorship, right? Which is like to have scientific impact, you often need to be a extraordinary emotional, like emotionally in tune person with the folks you're trying to influence. And so what comes so naturally to you is actually a super high stakes thing to other people. And so I wouldn't assume that Dario's more stressed out than you. These things are you'd be surprised how similar and small sometimes the problems are to you That some of the world's biggest, leaders are facing. And that's what I've learned from this class. The guest speakers are Sam, Satya, Jensen.Swyx [00:40:01]: AI Coachella.Anjney [00:40:02]: Yeah. It's AI Coachella, right? So we got to get all the headliners, and they're I'm very lucky that some of these people have either mentored me over the years or I've done business with them. And when you, take the performative stuff out and any assumptions you may have about these people that you read in the press or on Twitter, We're all just humans. We're all trying to get along. And what's so special about this moment is AI is forcing, like scaling, the bitter lesson is forcing a lot of people to revise their assumptions for how the world works and go back to first principles or go and educate themselves. So the kind of people I was, I won't name who this person is, but I was at an event last week in Texas and, ran to somebody who said, “Anjney, I came across the class. What do you think about real time action prediction models?” And I was, don't know how happy it made me feel when they asked me that question. I know they've done the work. They've challenged themselves. I'm, they didn't ask me, “What do you think of world models?” They said, “What do you think of n-”Swyx [00:41:04]: Real time action predictionAnjney [00:41:05]: “action, real time action prediction models?” World models, don't get me wrong, are cool and everything, but you and I both know that is a layer of abstraction that is sometimes not usefully precise enough. Right? Ours-Swyx [00:41:16]: There's like four different kinds of world models.Anjney [00:41:17]: Yes, exactly.Swyx [00:41:18]: We've done the part with general intuition, by the way, which is very focused on, -Anjney [00:41:22]: Oh, cool. Yes. I love Pim. Pim is great. And this is what I love about people who've done that level of work. They realize they're not in competition with people who the rest of the world thinks they're in competition with.Swyx [00:41:34]: Because they're not in the category, they're in the specific thing they're trying to do.Anjney [00:41:37]: They're focused on their mission, and they have a systems understanding of the bottleneck they're trying to solve. And when somebody else says, “I'm working on real time, action prediction models too,” Pim goes, “Oh, I love that person. I want, I can learn from them.” But the minute they're “Oh, that person's a world model person,” it's “like which type of world model person?” But mostly they're just trying to figure out if it's a waste of their time, because we don't have enough time. So, Pim, for example, is super, loves this other company I work with we've talked about called Black Forest Labs. And he's mentioned to me multiple times that he's so, He thinks what Flux is doing is really cool. Andy Blattman came by and spoke in the class. And what I find over and over again is for people who do the work, who can be usefully precise enough about like what is actually going on in the world of frontier research, The sense of camaraderie is still well and alive, but it gets lost sometimes when you have to like abstract The technical complexities in, business terms And then the VCs are “How are you different from that world model?” I'm going to say Where do I even start to explain this stuff? And then the misalignment creeps in.Leading vs. Winning in Frontier AISwyx [00:42:43]: This is good. Yeah, I think, people listening get a sense of, what it is like to operate at a real level, like yourself, rather than at, the journalist level, where you have to sort of put everyone in, a rough category and create a narrative of competition, and who's winning today, who's behind.Anjney [00:42:58]: It-- this idea of winning is so Weird to me.Swyx [00:43:03]: You do want to win. You want you want competitiveness.Anjney [00:43:06]: No, I think you want to lead.Swyx [00:43:07]: You want SOTA.Anjney [00:43:07]: No, I think you want to lead. Yes, so you want to push the frontier. You want to push the SOTA. You want to do something that hasn't been done before. You want to capture value, but you don't want to capture so much value that, people think you're unaligned with your mission or trying to do what's best for the world. You want to capture enough value that you can keep innovating, right? And I think that people want to lead, they don't really This idea of winning and losing, again, I love Jensen. He's a, he's a leader. The mindset that he talked about on Dwarkesh's podcast, right? He's “I didn't wake up with a loser mindset.” I think that was awesome, right? Because he's, he's an engineer. Dwarkesh has done the work. So there's at least-- even though the, to me, it was very obvious they're talking about the same thing, they just passed each other. They just had to basically, Jensen has this, five-layer cake abstraction of how the industry works. And Dwarkesh had, I think from that podcast, had more of, a pre-training, mid-training, post-training systems loop concept.Swyx [00:44:04]: It's just a factor of who he talks to, right? Again, it's very clear.Anjney [00:44:06]: It's the systems It's the abstraction, the mental models, the It's the whole-- Dude, so much of the problem in the world is reasoning by analogy. And then the assumptions that are held invisibly.Swyx [00:44:19]: Yeah, I've, I've said, this is actually the best time in human history for first principles thinkers. Because everything you think will happen is actually now coming true.Anjney [00:44:28]: Correct. And the venture capital community is, notorious for this, where people look-- In times of uncertainty, they, cling to axioms that ended up being true from the previous era, and they kind of like proclaim them with confidence as if they're truths, but they're not. And it's very important to see the distinction between a heuristic and an axiom. An axiom can be proven-Swyx [00:44:55]: Like from internal consistency point of viewAnjney [00:44:56]: With internal consistency. A heuristic is a way you kind of a shortcut. And my God, the number of people I have had to put up with over the last few years who proclaim-- use heuristics As axioms to judge people, to judge which companies are going to succeed or the number of people who are “Oh, yeah, Anthropic, they're just training models right now,” but this one continue.Swyx [00:45:22]: Because that's a B2B SaaS?Anjney [00:45:23]: Yeah, the, like Which over the fullness of time, if you squint at it, maybe. But the way you arrive there is so important that you can-- you just, you can dismiss people. Here's what happened, right? What happened is Anthropic basically achieved takeoff in October of last year. That training run-Swyx [00:45:41]: Whatever, three seven?Anjney [00:45:42]: I forget the numbers now, but whatever that checkpoint was-Swyx [00:45:45]: We saw the cognition.Anjney [00:45:46]: Yeah. Right? You probably-- The, to those of us in the community, especially once post-training was done and it was released in December-Swyx [00:45:52]: Yeah. Can I sneak a sneaky question in there? I don't know if you have a perspective, maybe you don't, I just The number one question is how did Anthropic crack coding, right? Because Claude One, Claude Two, okay, like it was part of it, but it wasn't a big deal. And the leading hypothesis, it's a lucky dice roll that was then compounded, right? Like it was like Mildly better, but then they saw it and they were “Okay, let's really invest.”How Anthropic Cracked CodingAnjney [00:46:17]: I had this very annoying teacher. I went to this boarding school called Rishi Valley in India, which is like this, bird preserve. It's like three hundred and fifty acres of bird preserve in rural India, and there was no technology for seven years. There was this teacher, I won't name them, but they would have this-- I hated it every time he said this to me. He was “Luck fa-favors the prepared mind,” which is like a common saying, but the way he delivered it, always grated me, ‘cause he was always I was always one of those kids who got, a good grade without trying very hard. ‘Cause like high middle school is not that hard if you, if you're generally, paying attention and so on. And there was this one time where I-- But then I would get an eighty percent grade, and he would keep pushing me to say “The reason you didn't get the ninety-five plus percent is because you're not that lucky.” And I would say, “What do you mean?” ‘Cause I would think that I deserved that grade, and I would sometimes argue with him. And he'd say, “You didn't have a prepared mind. If you want to get lucky again “ There was basically one time where I got like ninety-five or ninety-six on this, on this subject, and I, now that I felt entitled. I was “Okay, I'm going to keep doing this,” and I didn't. And then he was “Luck favors a prepared mind. You got lucky last time, but you got to stay prepared.” And I didn't understand what he meant. Now, as I'm older, I'm okay, these adults actually knew a thing or two. Anthropic has been the most prepared company for four years. And so then when the right, context data comes in, the right developers start sending in, the right context diffs, Sure, you could say you got lucky, but if you ask me, they're pr-pretty damn prepared with paranoia for like four years. And you have to remember, it was so hard for them to get going early on that they had to do so much more with so much less that you just have to be prepared to be so efficient.Swyx [00:48:06]: Yes. There's numbers on their burn compared to OpenAI. I've, I've written about it, but they are so much more efficient in their, in their tech stack.Anjney [00:48:14]: It's not even It's not funny.Swyx [00:48:14]: Not even close.Anjney [00:48:15]: Yeah. But it's so clear, right? Like how to output max for the world. They have been prepared, and you could call that luck, but Luck favors the prepared mind.Culture, Hardship, and Anthropic's P0Swyx [00:48:25]: This is one of those things that I was going over some of your old lectures and, you were data, people think it's a moat and actually it's culture and actually it's team Actually. And I, it's-- there's different levels of moats, and this is the ultimate one that determines everything else. Which you can then compoundAnjney [00:48:43]: You're saying culture is the ultimate moat? Yeah. But the thing about culture is it's very fragile. So moats, I don't think they're-- there's very few moats I found that are actually moats. They're-- It's, it's a nice concept, but in reality, you have to replenish your culture. Ben Horowitz was, the speaker in CS153 on Tuesday, and I asked him this question about the culture bottleneck in teams because, there are several AI teams-Swyx [00:49:09]: His book, Hard Things About Hard ThingsAnjney [00:49:11]: Hard Thing About Hard Things. But more concretely, there are so many AI labs today that have all the cash they need, they have all the compute they need, and they're still not able to ship anything SOTA. And then you start seeing people leave and so on, and my diagnosis, it's, is it's the culture. And so I asked him, Ben, they're-- He's been one of the most aggressive investors in AI labs. He goes back to this thing which resonates in my mind a lot. It-- When I used to work at a16z, I would, book a conference room, and right outside the conference room, which is closest to the toilet ‘cause it was the fastest way for me to go use the bathroom between Zoom meetings-Swyx [00:49:45]: Oh my God, I'll put maxing my toilet optimization. Okay, never mind.Anjney [00:49:48]: It was not healthy in hindsight, but maybe this is TMI. But anyway, outside that conference on the wall was this quote that was printed that said, “Culture is not a set of beliefs, it's a set of actions.” And it's by Bushido, is this, Japanese philosopher. And if you stop taking the actions that demonstrate the mission alignment to what you've said to your team and to your-- the world matters to you, then your culture starts to fray. So it's not actually a moat, I would say. It's a very brittle, fragile thing that requires daily tending to like a garden. But if you figure out the system to keep that garden tended, which I think ultimately comes down to knowing yourself ‘cause you most naturally, if you're authentic and so on, you'll naturally make trade-offs that seem effortless to you, but that reinforce your culture. And then That becomes this very hard thing for other people to catch up to. And at Anthropic, from day one, there was this mission like-- missionary like zeal and belief that, hey, these capabilities will scale. These systems are stochastic, not deterministic. There will be error bars, and until we crack interpretability, there's risk. And at some point, people will go-- stop using Claude just for coding. They'll use it in some mission-critical context where there's-- it'll throw off a bug, and then people are going to come blame them, and they want to be on the right side of history where they said, “Yes, this is a powerful technology. We think it's going to change the world, And we want to be very measured and scientific about the fact that, ‘Hey, guys, these are stats models, statistical models.' That's how statistics works.” ultimately, when you're training neural nets, it is just a statistical system. And I think that Belief that safety is important and that it might seem toy-like in the early days, and sometimes, you could say, “Anjney, they totally over-exaggerated the risk,” like two years ago when they said, “Let's not launch Claude One,” or whatever. Well, okay, maybe in hindsight, but hindsight is twenty/twenty. And at the time, they didn't know how that model would be used, and to them it felt existential if somebody came and said, “You weren't responsible. It-- This wrote a bug.” The liability associated with that is massive. So how do you prevent against that? Well, day in, day out, you say safety. And when you start deviating from that, you have the team hold you accountable, you have the world hold you accountable, and I think that becomes a moat over time. At some point, that moat will get challenged and so on, and then it become fragile. I hope it endures because that's the beauty of having founders run the show, ‘cause they can make really hard trade-offs to do mission alignment. The hardest part is in the earliest days when you don't have a group of people who are going through difficulty, stress, crisis together, then your culture doesn't get defined sharply enough, and that's what I'm worried about right now, is there's so much money going to these labs. There's no hardship. There's no-Swyx [00:52:50]: To anyone who knowsAnjney [00:52:51]: There's no to anyone who knows. And that, in hindsight, was a feature, not a bug for Anthropic. The number of people who said no, the number of people who said, “Sorry, we're all doing investors in OpenAI,” that is competitive difference. It forces you to really understand, what is the hill you want to die on at the expense of everything else. What's the P zero? And there, P zero from day one was coding. The reason, the mechanism system there was if we crack coding, Then we will crack AGI. Our mission is AGI. We want to get there safely. If we focus on codin

This episode breaks down gestational diabetes (new insulin resistance in pregnancy—not type 1 or 2), how insulin and glucose work, and why pregnancy naturally raises insulin resistance (but sometimes the pancreas can't keep up). It covers who's at higher risk (higher BMI, inactivity, prior GDM, certain ethnic groups, prior 9+ lb baby, PCOS, hypertension, heart disease) and how screening works at 24–28 weeks: the non-fasting 1-hour 50g test, followed by a fasting 3-hour 100g test if you “fail” (two elevated values = diagnosis). It explains why the test is a glucose “stress test,” alternatives like at-home fingersticks or CGM, and drink options (Glucola, Fresh Test, jelly beans). Finally, it outlines risks (C-section, shoulder dystocia, preeclampsia, baby hypoglycemia) and care differences between diet-controlled (deliver by 40+6) vs medication-controlled (NSTs/ultrasounds ~32 weeks, deliver ~39).00:00 Gestational Diabetes Overview00:52 How Insulin Works02:31 Risk Factors Explained03:20 Glucose Test Basics05:43 One Hour vs Three Hour07:06 At Home Monitoring Options09:42 Drink Alternatives13:52 Why It Matters Risks18:46 How Care Changes21:27 Final Recap Next Steps24:24 Closing Resources————

The GDM guys continue their talk from a couple of weeks ago about high level Dungeons & Dragons. This time they focus in on 4th and 5th editions of the game and why they break down at high levels. Enjoy!

Most people working on AI safety think without a massive effort AI systems will probably end up with goals catastrophically different from humanity's. Today's guest, Rohin Shah — head of AGI Safety and Alignment at Google DeepMind, and an AI safety researcher since 2017 — disagrees.“There is no particularly compelling argument that this is the thing that happens by default,” Rohin explains. “There's a lot of arguments that are suggestive that maybe it could happen, such that you should find it plausible. That's sufficient to justify a significant amount of effort into averting it, which is why I work in the area I do. But none of them rise to the level of, ‘I'm expecting this to happen by default.'”Take the worry that AIs will accidentally be trained to be deceptive. Sure, it's possible. But we're not running reinforcement learning over year-long trajectories — for now, we're running it over a week at most. The natural prediction is that models learn to grab short-term reward, not that they develop the ambitious long-horizon goals required for convergent power-seeking.What about current examples of models lying and scheming? Rohin has looked into the details, and most don't really resemble the thing we really fear: a competent AI pursuing an ambitious misaligned goal. Anthropic's “alignment faking” results, for instance, show a model trying to preserve its trained values against modification, which is arguably what it was trained to do.Rohin also expects we'll see problems coming. There's some generalisation risk at the point where AIs become powerful enough to actually take over, but the underlying challenges — overseeing superhuman systems, interpretability — are things we can iterate on now.Host Rob Wiblin pushes back on the case for AI optimism, and they also explore why current alignment success isn't strong evidence about superhuman systems, what it would actually take to change Rohin's mind, and where he thinks the doomers go wrong.Learn more, video, and full transcript: https://80k.info/rs26Check out our new book! https://80k.info/career-guideChapters:Who's Rohin Shah? (00:00:00)Rohin thinks we probably won't get catastrophic misalignment (00:00:49)Safety 'commitments' have severe limitations (00:10:38)Rohin's team doesn't have a veto and that's OK (00:27:36)Central banks are a promising model for regulating AI (00:33:34)'Pre-deployment evals' are overrated (for catastrophic risks) (00:37:41)Governance is likely a bigger bottleneck than alignment (00:43:55)Why isn't Rohin trying to pause AI progress? (00:51:44)We'll probably be able to read AI thoughts for years to come (00:54:17)Having to signal concern for safety can divert resources from actually making AI safer (01:09:51)A very underrated GDM paper (01:28:59)Google DeepMind's actual plan for building AGI safely (01:40:29)Why Rohin doubts the intelligence explosion is imminent (01:52:44)How external researchers can positively influence big AI companies (02:21:55)The roles GDM most needs to hire for (02:37:03)How Rohin stays positive (02:42:55)  This episode was recorded on December 4, 2025.Our production team includes:Video editors: Josh Alward, Dominic Armstrong, Jasper Luithlen, Milo McGuire, Luke Monsour, and Simon MonsourProducers: Elizabeth Cox and Nick StocktonCoordination and support: Katy Moore and Lou MoranCamera operator: Jeremy Chevillotte

The GDM guys talk about all the different editions of Dungeons & Dragons and how they each handle high level play. They also dive into the latest Wizards of the Coast fiasco and their digital downloads. Also a character death in Caverns of Thracia.

Outside of pregnancy, guidelines emphasize diabetes self-management education and support to facilitate informed decision making, self-care behaviors, problem solving, and active collaboration with health care professionals. This includes, in those with good health literacy, the concept of patient-led self-titration of basal insulin results which has data that it improves glycemic management compared with clinician-led titration for type 2 diabetes among nonpregnant adults. But what about for GDM? Can patient's self manage their BASAL insulin? In this episode, we will review a new RCT published in April 2026 in the Green Journal on this very subject. As novel as this is, it is not the first to report on this as it was also published (retrospective study in the UK) in 2022. This is a novel approach to insulin in GDM but there are some questions that remain. Listen in for details.1. Boonpattharatthiti K, Wechkunanukul K, Mayang N, et al . Comparison of Insulin Titration Strategies for Glycemic Control in Type 2 Diabetes: A Systematic Review and Network Meta-Analysis.Diabetes Care. 2025. 2. Valent, Amy M. DO, MCR; Barbour, Linda A. MD, MSPH. Insulin Management for Gestational and Type 2 Diabetes in Pregnancy. Obstetrics & Gynecology 144(5):p 633-647, November 2024. | DOI: 10.1097/AOG.00000000000056403. Wang, Xiao-Yu MD; Gabbe, Steven MD; Landon, Mark B. MD; Venkatesh, Kartik K. MD, PhD et al. Patient-Led Insulin Titration for Glycemic Management With Gestational Diabetes Mellitus: A Randomized Controlled Trial. Obstetrics & Gynecology 147(4):p 501-509, April 2026. 4. McGovern AP, Hirwa KD, Wong AK, et al. Patient-led rapid titration of basal insulin in gestational diabetes is associated with improved glycaemic control and lower birthweight. Diabet Med. 2022;39:e14926. doi: 10.1111/dme.14926

Modern medicine has come a long way in its fight against diabetes. We now have continuous glucose monitors (CGM) and automated insulin delivery (AIDs) systems. These have revolutionized patient care. The FDA has approved devices for use in pregnancy as “nonadjunctive use” (meaning they may be used alone), although capillary finger stick assessments are currently still considered the Gold Standard. While the most robust data in support of CGMs is for preexisting Type 1 DM (Class B or beyond) and Type 2, there is recent growing support for CGM use in GDM patients, although some limitations still apply. Listen in for details.1. Feig DS, et al; CONCEPTT Collaborative Group. Continuous glucose monitoring in pregnant women with type 1 diabetes (CONCEPTT): a multicentre international randomised controlled trial. Lancet. 2017 Nov 25;390(10110):2347-2359. doi: 10.1016/S0140-6736(17)32400-5. Epub 2017 Sep 15. Erratum in: Lancet. 2017 Nov 25;390(10110):2346. 2. Benhalima K, Durnwald C, Sweeting A et al.Application of continuous glucose monitoring and automated insulin delivery technologies for pregnant women with type 1, type 2, or gestational diabetes: an international consensus statementThe Lancet Diabetes & Endocrinology, 2025; 14, 157-1773. Salmen BM, Reurean-Pintilei D, Salmen T, Bohîlțea RE. Exploring Continuous Glucose Monitoring in Gestational Diabetes: A Systematic Review. Life (Basel). 2025 Aug 28;15(9):1369. doi: 10.3390/life15091369. PMID: 41010309; PMCID: PMC12470761.4. Wyckoff JA, Lapolla A, Asias-Dinh BD, et al.Preexisting Diabetes and Pregnancy: An Endocrine Society and European Society of Endocrinology Joint Clinical Practice Guideline. The Journal of Clinical Endocrinology and Metabolism. 20255. American Diabetes Association Professional Practice Committee for Diabetes*; 15. Management of Diabetes in Pregnancy: Standards of Care in Diabetes—2026. Diabetes Care 1 January 2026; 49 (Supplement_1): S321–S338. https://doi.org/10.2337/dc26-S0156. Burk J, Ross GP, Hernandez TL, Colagiuri S, Sweeting A. Evidence for improved glucose metrics and perinatal outcomes with continuous glucose monitoring compared to self-monitoring in diabetes during pregnancy. Am J Obstet Gynecol. 2025 Sep;233(3):162-175. doi: 10.1016/j.ajog.2025.04.010. Epub 2025 Apr 10. PMID: 40216177.7. Linder T, et al; GRACE study collaborative group. Glycaemic control and pregnancy outcomes with real-time continuous glucose monitoring in gestational diabetes (GRACE): an open-label, multicentre, multinational, randomised controlled trial. Lancet Diabetes Endocrinol. 2026 Jan;14(1):50-61. doi: 10.1016/S2213-8587(25)00288-8. Epub 2025 Nov 24. PMID: 41308662.8. Valent AM, et al. Real-Time Continuous Glucose Monitoring in Pregnancies With Gestational Diabetes Mellitus: A Randomized Controlled Trial. Diabetes Care. 2025 Sep 1;48(9):1581-1588. doi: 10.2337/dc25-0115. PMID: 40730104; PMCID: PMC12368369.9. Kusinski LC, et al. Continuous Glucose Monitoring Metrics and Pregnancy Outcomes in Women With Gestational Diabetes Mellitus: A Secondary Analysis of the DiGest Trial. Diabetes Care. 2025 Aug 19:dc250452. doi: 10.2337/dc25-0452. Epub ahead of print. PMID: 40828742; PMCID: PMC7618813.10. García-Moreno RM, et al. Efficacy of continuous glucose monitoring on maternal and neonatal outcomes in gestational diabetes mellitus: a systematic review and meta-analysis of randomized clinical trials. Diabet Med. 2022 Jan;39(1):e14703. doi: 10.1111/dme.14703. Epub 2021 Oct 13. PMID: 34564868.11. Amylidi-Mohr Set,.et al (DipGluMo): an open-label, single-centre, randomised, controlled trial. Lancet Diabetes Endocrinol. 2025 Jul;13(7):591-599. doi: 10.1016/S2213-8587(25)00063-4. Epub 2025 May 26. Erratum in: Lancet Diabetes Endocrinol. 2026 Mar;14(3):e6. doi: 10.1016/S2213-8587(25)00403-6. PMID: 40441173.

The following article of the Energy industry is: “Mexican Energy: Strategic Opportunities in a Transforming World” by Gamaliel Corral, CEO, GDM & MIP CINCO GAS. (AA1204)

Gestational diabetes (GDM) is one of the most common reasons families are advised to plan for an early birth. But what does the evidence actually say about induction for GDM? Does it lower the risk of Cesarean? Prevent big babies? Reduce stillbirth? Or does the timing matter more than the induction itself? In this episode, Dr. Rebecca Dekker and Dr. Morgan Richardson Cayama walk through the updated research on induction for gestational diabetes. You'll learn how outcomes differ before 39 weeks, between 39–40 weeks, and after 41 weeks, and why blood sugar control (diet-controlled versus medication-controlled GDM) can change the conversation entirely. They also review what major professional organizations recommend and discuss the role of extra fetal monitoring in the third trimester. Most importantly, they talk about informed consent, respectful maternity care, and how to navigate conversations if you're feeling pressure to schedule an induction. (00:02:40) Background & research update (00:05:34) What is GDM? Risks & induction rates (00:08:34) Research challenges & study limitations (00:15:36) Timing of birth: 38, 39, 40+ weeks (00:19:26) Big babies & health risks (00:24:27) Professional guidelines (ACOG, NICE, SOGC) (00:27:14) Birth before 41 weeks: common recommendation (00:27:54) Extra fetal monitoring in late pregnancy (00:32:49) Navigating pressure & informed consent View the full list of references here. Resources Read the updated Evidence on: Induction for Gestational Diabetes: ebbirth.com/inducingGDM Get the free respectful care handout: ebbirth.com/369 Grab your Pocket Guide to Labor Induction here. EBB 370 - Updated Evidence on Diagnosing Gestational Diabetes

As BMIs and weights increase across the US population, there have been increased calls for universal screening for existing DM at entrance to prenatal care, if under 20 weeks. Others, including the ACOG, prefer to screen early those with additional risk factors (like prior GDM HX, prior macrosomia, BMI >30, PCOS, first degree relative with diabetes, or age >40). In July 2024, the ACOG released its publication, “Screening for Gestational and Pregestational Diabetes in Pregnancy and Postpartum”. In this guidance, it states, “At this time, there are insufficient data to support the best screening modality for pregestational diabetes in pregnancy, but consideration can be made to use the same diagnostic criteria as for the nonpregnant population (A1c value 6.5 or higher, or fasting plasma glucose value 126 mg/dL or higher, or 2-hour plasma glucose value 200 mg/dL or higher during a 75-g OGTT, or random plasma glucose value 200 mg/dL or higher in patients with classic hyperglycemia symptoms)”. However, a new proposed protocol has been published in AJOG for early screening for DM in pregnancy. This also describes the differences in diagnosis and care for Standard GDM diagnosed at 24-28 weeks, vs a diagnosis of pregestational DM diagnosis made prior to 20-weeks vs “early” GDM also diagnosed under 20 weeks of gestation. Listen in for details. 1. McLaren, Rodney et al.nA Proposed Classification of Diabetes Mellitus in PregnancyAmerican Journal of Obstetrics & Gynecology, Volume 0, Issue 0. Epub Feb 2, 2026; https://www.ajog.org/article/S0002-9378(26)00061-X/fulltext2. ACOG Clinical Practice Update: Screening for Gestational and Pregestational Diabetes in Pregnancy and Postpartum; July 2024; https://journals.lww.com/greenjournal/abstract/2024/07000/acog_clinical_practice_update__screening_for.34.aspx3. Simmons, David et al. “Treatment of Gestational Diabetes Mellitus Diagnosed Early in Pregnancy.” The New England journal of medicine vol. 388,23 (2023): 2132-2144. doi:10.1056/NEJMoa2214956

This show has been flagged as Clean by the host. Create a Linux kiosk at your library Start without a guest account The first few steps of this process don't actually require a guest user directory to exist, so do NOT create your guest user account yet. However, you do need to choose what your guest user account is going to be called. A reasonable account name for Don's purposes is libraryguest. On my personal computer I call my guest account guestaccount, and I've used kioskguest on some installations. I avoid just the name “guest” because in modern computing the term “guest” gets used in a few other ways (such as a “guest operating system” in a virtual environment), and it's just easier to find something unique in logs. Choose a unique name for you guest account, but don't create it yet. For this article, I'm using libraryguest. Create the PostSession script By default, GDM recognises several states: Init, PostLogin, PreSession, and PostSession. Each state has a directory located in /etc/gdm. When you place a shell script called Default in one of those directories, GDM runs the script when it reaches that state. To trigger actions to clean up a user's environment upon logout, create the file /etc/gdm/PostSession/Default. You can add whatever actions you want to run upon logout to the Default script. In the case of Don's library, we wanted to clear everything from the guest's home directory, including browser history, any LibreOffice files or GIMP files they may have created, and so on. It was important that we limited the very drastic action of removing all user data to just the guest user. We didn't want the admin's data to be erased upon logout, so whatever rule we added to /etc/gdm/PostSession/Default had to be limited to the guest user. Here's what we came up with: #!/usr/bin/sh echo "$USER logged out at `date`" >> /tmp/PostSession.log if [ "X$USER" = "Xlibraryguest" ]; then rm -rf "$HOME" fi exit 0 The first line is for logging purposes. The /tmp directory gets cleared out on most distributions automatically, so we weren't worried about creating a file that'll grow forever and eventually crash the computer. If your distribution of choice doesn't clean out /tmp automatically, create a cron job to do that for you. GDM knows what user triggered the logout process, so the if statement verifies that the user logging out is definitely the libraryguest user (that's the literal name of the user we created for library patrons).Note that the whitespace around the square brackets is important, so be precise when typing! As long as it is libraryguest, then the script removes the entire user directory ($HOME). That can be extremely dangerous if you make a mistake, so do thorough testing on a dummy system before implementing a script like this! If you get a condition wrong, you could erase your entire home directory upon logout. In this example, I've successfully limited the rm command to a logout action performed by user libraryguest. The entire /home/libraryguest directory is erased, and the computer returns to the GDM login screen. When a new user logs in, a fresh directory is created for the user. You can put any number of commands in your script, of course. You don't have to erase an entire directory. If all you really want to do is clear browser history and any stray data, then you can do that instead. If you need to copy specific configuration files into the environment, you can do that during the PreSession state. Just be sure to test thoroughly before committing your creation to your users! What happens when the guest doesn't log out At this point, the computer erases all of the user's data when the user logs out, but a reboot or a shutdown is different to a logout. GDM doesn't enter a PostSession state after a reboot signal has been received, even if the reboot occurs during an active GDM session. The easiest and safest way to erase an entire home directory when there's a cut to system power is to use a temporary RAM filesystem (tmpfs) to house the data in the first place. If the systems you're configuring have 8 GB or more, and the system is exclusively used as a guest computer, you can probably afford to use RAM as the guest's home directory. If your system doesn't have a lot of RAM, then you can use the systemd work-around in the next section. Assuming you have the RAM to spare, and that your systems are supported by a backup power supply, you can add a tmpfs entry in /etc/fstab. In this example, my tmpfs is mounted to /home/libraryguest and is just 2 GB: tmpfs /home/libraryguest tmpfs rw,nosuid,nodev,size=2G 0 0 That's plenty of space for some Internet browsing and even a few LibreOffice documents to be saved while a user works. Mount the new volume: $ sudo mount /home/libraryguest Next, you must create the libraryguest user manually in a terminal.The useradd command creates user profiles: $ sudo useradd --home-dir /home/libraryguest libraryguest useradd: warning: the home directory /home/libraryguest/ already exists. useradd: Not copying any file from skel directory into it. Because you've already created a location for the home directory, you do get a warning after creating the user. It's only a warning, not a fatal error, and the guest account is automatically populated later. Create a password for the new user: $ sudo passwd libraryguest That's it! You've created a guest account that refreshes with every logout and every reboot. You can skip over the next section of this article. Using systemd targets instead of a ramdisk Assuming you can't create a ramdisk for temporary user data, you can instead create a systemd service that runs a script when the reboot, poweroff, and multi-user targets are triggered: [Unit] Description=Kiosk cleanup [Service] Type=oneshot ExecStart=/usr/local/bin/kiosk-cleanup.sh [Install] WantedBy=poweroff.target reboot.target multi-user.target Save the file to /etc/systemd/system/kioskmode.service and then enable it: $ sudo systemctl enable --now kioskmode The script, like the GDM script, removes the libraryguest directory. Unlike GDM script, this one must also recreate an empty home directory and grant it user permissions: #!/usr/bin/bash rm -rf /home/libraryguest mkdir /home/libraryguest chown -R libraryguest:libraryguest /home/libraryguest Grant the script itself permission to run: $ sudo chmod +x /usr/local/bin/kiosk-cleanup.sh Now the libraryguest user data is erased after: Logout Reboot Shutdown Startup Essentially, no matter how the computer loses its session or its power, the libraryguest account starts fresh when a new session is started. Security and privacy Using systemd to erase data at shutdown and startup isn't strictly as secure as using a temporary ramdisk for all user data. Should the computer lose power suddenly, all saved user data in the libraryguest account is present during the next boot. Of course, it's erased as soon as multi-user.target is called by systemd, but it is technically possible to interrupt the boot process and mine for data. You must use full drive encryption to protect data from being discovered by an interrupted boot sequence. Why not just use xguest On many Linux distributions, the xguest package is designed to provide the Guest account, which resets after each logout. It was an extremely useful package that I installed on every machine I owned, because it's handy to be able to let friends use my computer without risking them making a mess of my home directory. Lately, it seems that xguest is failing to launch a desktop, however, presumably because it relies on X11. If xguest works for you in your tests, then you may want to use it instead of the solution I've presented here. My solution offers a lot of flexibility, thanks to GDM's autodetection of session states. Kiosks in libraries Privacy and personal information is more important than ever. Regardless of how you setup a kiosk for your library, you have an obligation to your users to keep them informed of how their data is being stored. This goes both ways. Users need to know that their data is destined to be erased as soon as they log out, and also they deserve to be assured that their data is not retained. However, it's also your responsibility to admit that glitches and exceptions could occur. Users need to understand that the computer they're using are public computers on a public network. Encryption is being used for traffic and for data storage, but you cannot guarantee absolute privacy. As long as everyone understands the arrangement, everyone can compute with confidence. Linux, GDM, and systemd are great tools to help libraries create a sustainable, robust, honest, and communal computing platform. Show notes taken from https://www.both.org/?p=13327

As 2025 comes to an end, guest host Dr. Sara Ailshire turns the tables and interviews Dr. Rebecca Dekker about the biggest childbirth trends, lessons, and breakthroughs of 2025, and what exciting changes are coming to EBB in 2026.   Together, Sara and Rebecca dive into the shifting landscape of birth: the unprecedented rise in labor inductions, how AI is complicating the search for evidence-based information, changes in doula access and Medicaid coverage, and how politics continues to shape pregnancy and postpartum care. They walk through the most impactful EBB research updates of the year—including new evidence on vitamin K, gestational diabetes testing, induction timing, big babies, and respectful maternity care—and reflect on the episodes that resonated most with our global community.   Rebecca also opens up about what she personally learned this year, including how unresolved childhood trauma impacted her own labor years ago, and how that insight is shaping her thinking about the emotional and spiritual dimensions of birth. Plus, Rebecca reveals a major new direction for Evidence Based Birth in 2026 that could transform hospital birth culture around the world and bring evidence-based care to thousands more families.   Want to provide input on EBB's new direction? Fill out this survey here!   (02:12) The #1 trend of 2025: inductions everywhere (03:50) How AI is reshaping (and complicating) birth information (07:51) Doula coverage, Medicaid changes, and fewer parents seeking childbirth education (11:55) Miscarriage care, politics, and the impact of Dobbs (13:42) Biggest EBB research updates: vitamin K, GDM, and more (21:40) The new Respectful Maternity Care handout (22:21) The new "big baby" trial and why it likely won't shift U.S. practice (25:37) The top five EBB podcast episodes of the year (32:58) Highlights from the 2025 EBB Conference & Summer School (41:22) How trauma shaped Rebecca's own labor (53:50) The big reveal: what's coming for EBB in 2026   Resources Vitamin K Signature Article (Updated 2025): ebbirth.com/vitamink Gestational Diabetes Signature Article (Updated): ebbirth.com/gdm Get the Respectful Maternity Care Free Handout: ebbirth.com/RMC Sign up for the Big Baby Signature Training for Pro Members: ebbirth.com/classes Get the My Doula Visit Workbook: ebbirth.com/doula-workbook/   Referenced EBB Episodes EBB 349 – An L & D Nurse's Advice for Advocating in the Birth Room with Trish Ware the Labor Nurse Mama EBB 357 – Making Decisions about Elective Induction of Labor with Dr. Ann Peralta & Kari Radoff, CNM, from Partner to Decide EBB 377 – Medicaid Coverage for Doula Care with Amy Chen, Senior Attorney at the National Health Law Program EBB 352 – Calming Breathing Techniques for Pregnancy with Dr. Shilpa Babbar, Obstetrician and Maternal Fetal Medicine Specialist EBB 343 – Top Ten Evidence-Based Strategies for Lowering the Risk of Cesarean EBB 347 - Updated Evidence on Vitamin K EBB 350 – Surviving a Long Antepartum Hospital Stay and Preparing for a Scheduled Cesarean with Krista DeYoung, EBB Childbirth Class Graduate EBB 372 – Comfort Measures and a 41-Week Induction with Hopey Fink and Ben Levin, EBB Childbirth Class Graduates   EBB Doula Trainer Rewards Lorie Michaels, BirthPro Advanced Doula Training: birthpro.org Lorenda Lewis, Healing with Dignity: healingwithdignity.com Heather McCullough, HMBirth: hmbirth.com Heather Christine Struwe, Community Aware Birthworker: communityawarebirthworker.com Charlotte Shilo-Goudeau, Community Birth Companion: communitybirthcompanion.org Naima Beckles, For Your Birth: foryourbirth.com Leiko Hidaka, Leiko Hidaka: leikohidaka.com Ruth Kraft, Birth Professional International: birthprofessionalinternational.com Jennifer Anderson, Birth Fusion: birthfusion.com Chanté Perryman, Baby Dreams Maternity Concierge: babydreamsmc.com   For more information about Evidence Based Birth® and a crash course on evidence based care, visit www.ebbirth.com. Follow us on Instagram and YouTube! Ready to learn more? Grab an EBB Podcast Listening Guide or read Dr. Dekker's book, "Babies Are Not Pizzas: They're Born, Not Delivered!" If you want to get involved at EBB, join our Professional membership (scholarship options available) and get on the wait list for our EBB Instructor program. Find an EBB Instructor here, and click here to learn more about the EBB Childbirth Class.    

We often think of Large Language Models (LLMs) as all-knowing, but as the team reveals, they still struggle with the logic of a second-grader. Why can't ChatGPT reliably add large numbers? Why does it "hallucinate" the laws of physics? The answer lies in the architecture. This episode explores how *Category Theory* —an ultra-abstract branch of mathematics—could provide the "Periodic Table" for neural networks, turning the "alchemy" of modern AI into a rigorous science.In this deep-dive exploration, *Andrew Dudzik*, *Petar Velichkovich*, *Taco Cohen*, *Bruno Gavranović*, and *Paul Lessard* join host *Tim Scarfe* to discuss the fundamental limitations of today's AI and the radical mathematical framework that might fix them.TRANSCRIPT:https://app.rescript.info/public/share/LMreunA-BUpgP-2AkuEvxA7BAFuA-VJNAp2Ut4MkMWk---Key Insights in This Episode:* *The "Addition" Problem:* *Andrew Dudzik* explains why LLMs don't actually "know" math—they just recognize patterns. When you change a single digit in a long string of numbers, the pattern breaks because the model lacks the internal "machinery" to perform a simple carry operation.* *Beyond Alchemy:* deep learning is currently in its "alchemy" phase—we have powerful results, but we lack a unifying theory. Category Theory is proposed as the framework to move AI from trial-and-error to principled engineering. [00:13:49]* *Algebra with Colors:* To make Category Theory accessible, the guests use brilliant analogies—like thinking of matrices as *magnets with colors* that only snap together when the types match. This "partial compositionality" is the secret to building more complex internal reasoning. [00:09:17]* *Synthetic vs. Analytic Math:* *Paul Lessard* breaks down the philosophical shift needed in AI research: moving from "Analytic" math (what things are made of) to "Synthetic" math [00:23:41]---Why This Matters for AGIIf we want AI to solve the world's hardest scientific problems, it can't just be a "stochastic parrot." It needs to internalize the rules of logic and computation. By imbuing neural networks with categorical priors, researchers are attempting to build a future where AI doesn't just predict the next word—it understands the underlying structure of the universe.---TIMESTAMPS:00:00:00 The Failure of LLM Addition & Physics00:01:26 Tool Use vs Intrinsic Model Quality00:03:07 Efficiency Gains via Internalization00:04:28 Geometric Deep Learning & Equivariance00:07:05 Limitations of Group Theory00:09:17 Category Theory: Algebra with Colors00:11:25 The Systematic Guide of Lego-like Math00:13:49 The Alchemy Analogy & Unifying Theory00:15:33 Information Destruction & Reasoning00:18:00 Pathfinding & Monoids in Computation00:20:15 System 2 Reasoning & Error Awareness00:23:31 Analytic vs Synthetic Mathematics00:25:52 Morphisms & Weight Tying Basics00:26:48 2-Categories & Weight Sharing Theory00:28:55 Higher Categories & Emergence00:31:41 Compositionality & Recursive Folds00:34:05 Syntax vs Semantics in Network Design00:36:14 Homomorphisms & Multi-Sorted Syntax00:39:30 The Carrying Problem & Hopf FibrationsPetar Veličković (GDM)https://petar-v.com/Paul Lessardhttps://www.linkedin.com/in/paul-roy-lessard/Bruno Gavranovićhttps://www.brunogavranovic.com/Andrew Dudzik (GDM)https://www.linkedin.com/in/andrew-dudzik-222789142/---REFERENCES:Model:[00:01:05] Veohttps://deepmind.google/models/veo/[00:01:10] Geniehttps://deepmind.google/blog/genie-3-a-new-frontier-for-world-models/Paper:[00:04:30] Geometric Deep Learning Blueprinthttps://arxiv.org/abs/2104.13478https://www.youtube.com/watch?v=bIZB1hIJ4u8[00:16:45] AlphaGeometryhttps://arxiv.org/abs/2401.08312[00:16:55] AlphaCodehttps://arxiv.org/abs/2203.07814[00:17:05] FunSearchhttps://www.nature.com/articles/s41586-023-06924-6[00:37:00] Attention Is All You Needhttps://arxiv.org/abs/1706.03762[00:43:00] Categorical Deep Learninghttps://arxiv.org/abs/2402.15332

Pedro Domingos, author of the bestselling book "The Master Algorithm," introduces his latest work: Tensor Logic - a new programming language he believes could become the fundamental language for artificial intelligence.Think of it like this: Physics found its language in calculus. Circuit design found its language in Boolean logic. Pedro argues that AI has been missing its language - until now.**SPONSOR MESSAGES START**—Build your ideas with AI Studio from Google - http://ai.studio/build—Prolific - Quality data. From real people. For faster breakthroughs.https://www.prolific.com/?utm_source=mlst—cyber•Fund https://cyber.fund/?utm_source=mlst is a founder-led investment firm accelerating the cybernetic economyHiring a SF VC Principal: https://talent.cyber.fund/companies/cyber-fund-2/jobs/57674170-ai-investment-principal#content?utm_source=mlstSubmit investment deck: https://cyber.fund/contact?utm_source=mlst—**END**Current AI is split between two worlds that don't play well together:Deep Learning (neural networks, transformers, ChatGPT) - great at learning from data, terrible at logical reasoningSymbolic AI (logic programming, expert systems) - great at logical reasoning, terrible at learning from messy real-world dataTensor Logic unifies both. It's a single language where you can:Write logical rules that the system can actually learn and modifyDo transparent, verifiable reasoning (no hallucinations)Mix "fuzzy" analogical thinking with rock-solid deductionINTERACTIVE TRANSCRIPT:https://app.rescript.info/public/share/NP4vZQ-GTETeN_roB2vg64vbEcN7isjJtz4C86WSOhw TOC:00:00:00 - Introduction00:04:41 - What is Tensor Logic?00:09:59 - Tensor Logic vs PyTorch & Einsum00:17:50 - The Master Algorithm Connection00:20:41 - Predicate Invention & Learning New Concepts00:31:22 - Symmetries in AI & Physics00:35:30 - Computational Reducibility & The Universe00:43:34 - Technical Details: RNN Implementation00:45:35 - Turing Completeness Debate00:56:45 - Transformers vs Turing Machines01:02:32 - Reasoning in Embedding Space01:11:46 - Solving Hallucination with Deductive Modes01:16:17 - Adoption Strategy & Migration Path01:21:50 - AI Education & Abstraction01:24:50 - The Trillion-Dollar WasteREFSTensor Logic: The Language of AI [Pedro Domingos]https://arxiv.org/abs/2510.12269The Master Algorithm [Pedro Domingos]https://www.amazon.co.uk/Master-Algorithm-Ultimate-Learning-Machine/dp/0241004543 Einsum is All you Need (TIM ROCKTÄSCHEL)https://rockt.ai/2018/04/30/einsum https://www.youtube.com/watch?v=6DrCq8Ry2cw Autoregressive Large Language Models are Computationally Universal (Dale Schuurmans et al - GDM)https://arxiv.org/abs/2410.03170 Memory Augmented Large Language Models are Computationally Universal [Dale Schuurmans]https://arxiv.org/pdf/2301.04589 On the computational power of NNs [95/Siegelmann]https://binds.cs.umass.edu/papers/1995_Siegelmann_JComSysSci.pdf Sebastian Bubeckhttps://www.reddit.com/r/OpenAI/comments/1oacp38/openai_researcher_sebastian_bubeck_falsely_claims/ I am a strange loop - Hofstadterhttps://www.amazon.co.uk/Am-Strange-Loop-Douglas-Hofstadter/dp/0465030793 Stephen Wolframhttps://www.youtube.com/watch?v=dkpDjd2nHgo The Complex World: An Introduction to the Foundations of Complexity Science [David C. Krakauer]https://www.amazon.co.uk/Complex-World-Introduction-Foundations-Complexity/dp/1947864629 Geometric Deep Learninghttps://www.youtube.com/watch?v=bIZB1hIJ4u8Andrew Wilson (NYU)https://www.youtube.com/watch?v=M-jTeBCEGHcYi Mahttps://www.patreon.com/posts/yi-ma-scientific-141953348 Roger Penrose - road to realityhttps://www.amazon.co.uk/Road-Reality-Complete-Guide-Universe/dp/0099440687 Artificial Intelligence: A Modern Approach [Russel and Norvig]https://www.amazon.co.uk/Artificial-Intelligence-Modern-Approach-Global/dp/1292153962

Gestational diabetes is one of the most common and challenging complications seen in pregnancy, and new guidelines are changing how GPs should approach its screening and management. In this HealthCert GP Update, Dr Simone Gonzo unpacks the Australasian Diabetes in Pregnancy Society (ADIPS) 2025 Consensus Recommendations, providing a clear, practical overview tailored for busy GPs. What you will learn How to distinguish overt diabetes from GDM. When and how to apply early screening protocols for high-risk patients. Updated diagnostic thresholds and what they mean in day-to-day practice. How these changes may influence your management decisions and patient outcomes. Next steps in your learning journey

video: https://youtu.be/FK81IHWIPPE Comment on the TWIL Forum This week in Linux, we've got some new distro releases from Linux Mint with Linux Mint 22.2, AerynOS, and the Linux from Scratch project. Also, I am going to introduce you to a brand new tool called WinBoat, their goal is to make it easy to run Windows apps on Linux. Then we got an update related Mozilla's deal with Google for the default search engine in Firefox and it looks like the GNOME team have re-enabled X11 Support by default in GDM for GNOME 49. All of this and more on This Week in Linux, the weekly news show that keeps you up to date with what's going on in the Linux and Open Source world. Now let's jump right into Your Source for Linux GNews! Sponsored by Sandfly Security: the revolutionary agentless platform designed for Linux. Visit https://thisweekinlinux.com/sandfly to experience security that's not just effective but gives you peace of mind. No agents. No downtime. Just cutting-edge protection. Download as MP3 Support the Show Become a Patron = tuxdigital.com/membership Store = tuxdigital.com/store Chapters: 00:00 Intro 00:50 Linux Mint 22.2 05:39 Linux from Scratch 12.4 09:38 WinBoat: Run Windows Apps on Linux 13:21 Sandfly Security, agentless Linux security 15:44 GNOME 49 Release Candidate Re-Enables X11 Support by Default in GDM 18:05 AerynOS August 2025 Update 20:15 Kazeta: Old Console OS in the Modern Era 22:42 Firefox to keep Google Search Deal 25:02 Valve "STEAM FRAME", is it New Hardware? 26:51 Outro Links: Linux Mint 22.2 https://blog.linuxmint.com/?p=4881 https://www.linuxmint.com/rel_zara_whatsnew.php Linux from Scratch 12.4 https://www.linuxfromscratch.org/lfs/view/12.4/ https://www.linuxfromscratch.org/blfs/view/12.4/ WinBoat: Run Windows Apps on Linux https://www.winboat.app/ https://github.com/TibixDev/winboat https://www.gamingonlinux.com/2025/09/winboat-is-a-new-linux-app-to-run-windows-apps-with-seamless-integration/ Sandfly Security, agentless Linux security https://thisweekinlinux.com/sandfly https://destinationlinux.net/ GNOME 49 Release Candidate Re-Enables X11 Support by Default in GDM https://discourse.gnome.org/t/gnome-49-rc-released/31234 https://download.gnome.org/teams/releng/49.rc/NEWS https://www.omgubuntu.co.uk/2025/09/gnome-49-reenables-x11-session-support-in-gdm AerynOS August 2025 Update https://aerynos.com/blog/2025/08/31/august-2025-project-update/ Kazeta: Old Console OS in the Modern Era https://kazeta.org/ https://www.gamingonlinux.com/2025/09/chimeraos-dev-announced-kazeta-a-new-linux-os-aimed-at-recreating-a-classic-console-experience/ Firefox to keep Google Search Deal https://www.omgubuntu.co.uk/2025/09/google-antitrust-ruling-firefox-search-deal Valve "STEAM FRAME", is it New Hardware? https://uspto.report/TM/99370857 https://uspto.report/TM/99370861 https://www.gamingonlinux.com/2025/09/new-valve-trademark-for-steam-frame-looks-like-were-getting-new-hardware/ https://www.pcgamer.com/hardware/gaming-pcs/valve-applies-to-use-steam-frame-as-a-trademark-for-a-new-console-as-speculation-over-a-mythical-next-gen-half-life-game-continues/ Support the show https://tuxdigital.com/membership https://store.tuxdigital.com/

Gestational diabetes (GDM) is one of the most common health issues during pregnancy, and diagnosing it is more complicated than you might think. In this episode, Dr. Dekker is joined by EBB Research Team member Dr. Morgan Richardson Cayama to cover the newly updated evidence on how GDM is diagnosed. They walk through the physiology behind GDM, current testing methods, and why there's still international disagreement about how to screen. Together, they examine the results of large randomized trials comparing the one-step and two-step screening methods, the research on early screening with hemoglobin A1C, and the evidence on alternatives to the Glucola drink, including candy and home blood sugar monitoring. They also review the risks of skipping screening entirely, and how weight bias and other systemic factors can impact diagnosis and care.   (02:28) What is Gestational Diabetes and Why Is It So Common? (06:30) Risk Factors, Size Bias, and the Role of Race and Ethnicity (10:40) Why We Screen and the Origins of the Controversy (13:17) Comparing the One-Step and Two-Step Methods (19:55) What New Research Says About Health Outcomes (23:45) Should We Screen for GDM Earlier in Pregnancy? (28:11) Can Hemoglobin A1C Replace the Glucola Drink? (32:44) Alternatives: Candy, Food, and Home Monitoring (40:04) What International Guidelines Recommend (43:07) Declining GDM Testing: What the Evidence Shows (47:47) Is Sperm Linked to Gestational Diabetes Risk? (51:29) Takeaways and the Future of GDM Diagnosis   Resources Download the free two-page handout in English or Spanish [NEED LINK] Explore Real Food for Gestational Diabetes by Lily Nichols: realfoodforgd.com For a full list of resources, visit ebbirth.com/inducinggdm For more information about Evidence Based Birth® and a crash course on evidence based care, visit www.ebbirth.com. Follow us on Instagram and YouTube! Ready to learn more? Grab an EBB Podcast Listening Guide or read Dr. Dekker's book, "Babies Are Not Pizzas: They're Born, Not Delivered!" If you want to get involved at EBB, join our Professional membership (scholarship options available) and get on the wait list for our EBB Instructor program. Find an EBB Instructor here, and click here to learn more about the EBB Childbirth Class.

According to Purdue University's Ag Economy Barometer survey in July, high input costs and lower crop and livestock prices are the top two concerns on farmers mind at the moment. For agbioscience innovators, it's a critical more than ever to consider moves that can deliver maximum value to the farmer. Dave Pugh, CFO of AgReliant Genetics, joins us as GDM recently announced its agreement to acquire the company. We get into:  Dave's background in computer science and risk management + what drew him to agbioscience Massive validation in the form of GDM's agreement to acquire AgReliant Genetics and what that means for the farmer How Dave thinks through the investment of innovation in agbioscience in a time of uncertainty for farmers and companies alike The AgReliant Genetics portfolio heading into acquisition and its strengths in corn genetics His perspective on the relationship between seed brands and customer Dave's finance background and role as CFO – in a time of uncertainty – planning for oversight of what's ahead and how to stay innovative in the process What has him most excited about what's ahead with GDM  The following conversation discusses the recent announcement that GDM is seeking to acquire AgReliant Genetics. The transaction is subject to regulatory approvals in the United States and other customary closing conditions and approvals. Until the necessary approvals and closing conditions are obtained and satisfied and the transaction has closed, GDM and AgReliant Genetics will continue to operate as independent entities, maintaining their current business routines and commercial structures. 

This is perhaps the most epic news recap we have ever encountered on Agbioscience! So much so, we had to go back and record audio after we thought we were done. We get into the USDA's announcement on re-organization and Indiana's inclusion as a regional hub as part of that, AgReliant Genetics' acquisition by GDM, BiomEdit's product pipeline milestones, Series B funding and new leadership additions, and a big partnership announcement and honor for Corteva Agriscience. We also provide an update on the AgriNovus CEO search and when you can expect to learn more information. Don't forget! AgriNovus Quadrant is August 20 -- Register here: https://agrinovusindiana.com/quadrant/USDA Reorganization Announcement + MemorandumISDA's Don Lamb Talks to Hoosier Ag Today on USDA ReorganizationAgReliant Genetics Announces Acquisition Agreement by GDMBiomEdit Announces Series B Raise, Advances Product to Final PhaseAgFunder News Talks with BiomEdit on Designer ProbioticsCountryMark Completes $100M Diesel Expansion ProjectPhytoform and Corteva Partner on AI to Boost Disease Resistance in Corn

This is perhaps the most epic news recap we have ever encountered on Agbioscience! So much so, we had to go back and record audio after we thought we were done. We get into the USDA's announcement on re-organization and Indiana's inclusion as a regional hub as part of that, AgReliant Genetics' acquisition by GDM, BiomEdit's product pipeline milestones, Series B funding and new leadership additions, and a big partnership announcement and honor for Corteva Agriscience. We also provide an update on the AgriNovus CEO search and when you can expect to learn more information. Don't forget! AgriNovus Quadrant is August 20 -- Register here: https://agrinovusindiana.com/quadrant/USDA Reorganization Announcement + MemorandumISDA's Don Lamb Talks to Hoosier Ag Today on USDA ReorganizationAgReliant Genetics Announces Acquisition Agreement by GDMBiomEdit Announces Series B Raise, Advances Product to Final PhaseAgFunder News Talks with BiomEdit on Designer ProbioticsCountryMark Completes $100M Diesel Expansion ProjectPhytoform and Corteva Partner on AI to Boost Disease Resistance in Corn

We've all heard about the infamous sugary drink

Gestational diabetes (GDM) can feel overwhelming, and for many women, it comes with confusion, fear, or guilt. But a diagnosis doesn't mean you've done anything wrong, and it certainly doesn't mean you're powerless.In this episode, we're joined by Boob to Food clinic dietitian and nutritionist Niki Mohtat to explore what GDM actually is, why it happens, and how you can manage it with confidence through nourishing food, supportive lifestyle tweaks, and the right guidance. Niki Mohtat is a dietitian and nutritionist with a passion for supporting women through preconception, pregnancy, and postpartum. Her interest in prenatal nutrition began with her own pregnancies, where she saw firsthand how much of the information available to the public was outdated or unhelpful. This inspired her to dedicate her career to providing evidence-based, individualised, and practical nutrition guidance, so women can feel confident nourishing themselves and their growing baby. Niki has completed advanced training through The Institute for Prenatal Nutrition Mentorship Program under Lily Nichols, a highly sought-after and competitive prenatal certification program. She offers consultations through the Boob to Food online clinic.In this episode, we discuss:What gestational diabetes really is and why it happens during pregnancyWhat the oral glucose tolerance test involves and its limitationsHow to approach food without fear, guilt, or perfectionismThe role of protein, fats, and carbs (and why carbs aren't the enemy)Tips for managing fasting blood glucose levelsThe connection between GDM and future health risksSimple strategies for postpartum and long-term wellbeing... and so much moreResources mentioned in this episode:Boob to Food Online ClinicOur earlier Boob to Food episode on preconception nutritionToday's episode is brought to you by Haakaa. Haakaa is a family-owned New Zealand brand committed to making motherhood simpler, easier, and greener. From their iconic breast pumps to their fresh food feeders and silicone freezer trays, Haakaa's range of safe, sustainable and non-toxic baby products are favourites in both of our homes. Whether you're breastfeeding, introducing solids, or prepping meals for your toddler, Haakaa offers practical solutions that support you every step of the way.You can use the code BOOBTOFOOD for 10% off your order at www.haakaa.co.nzFollow us on instagram @boobtofood to stay up to date with all the podcast news, recipes and other content that we bring to help make meal times and family life easier.Visit www.boobtofood.com for blogs and resources, to book an appointment with one of our amazing practitioners and more.Presented by Luka McCabe and Kate HolmTo get in touch please email podcast@boobtofood.com

In this episode of The Birth Lounge podcast, host HeHe discusses one of the most requested topics, Gestational Diabetes Mellitus (GDM), with midwife Melissa Chappell. Melissa, who owns Utah Birth Suites and founded Songbird Maternity, offers a holistic view on women's health and is a staunch advocate for informed consent and patient autonomy. The conversation dives deep into what GDM is, how it occurs, the importance of testing, and why the typical 50-gram glucose challenge may not always reflect reality. The duo also covers alternative testing methods, including continuous glucose monitoring, diet and exercise's role, and the impact of GDM on the baby. Melissa sheds light on the myths and truths about GDM, the implications of controlled versus uncontrolled GDM, and the risks associated with traditional medical practices, such as induction. Listeners will walk away feeling informed and empowered to engage in confident discussions with their healthcare providers. Tune in to get evidence-based insights, especially focusing on holistic and patient-centered care approaches for managing gestational diabetes.   00:00 Introduction to Gestational Diabetes 01:27 Meet Our Expert Guest: Melissa Chappell 02:33 Understanding Gestational Diabetes Mellitus (GDM) 03:03 Testing and Alternatives for Gestational Diabetes 03:53 Impact of Gestational Diabetes on Mother and Baby 10:16 Historical Perspective and Current Statistics 12:42 Challenges with Current Testing Methods 16:26 Managing Gestational Diabetes: Diet and Monitoring 24:07 Risks and Misconceptions about Big Babies 31:37 Pitocin Use and Its Implications 39:13 Increased Medical Interventions in Pregnancy 40:17 The Impact of Glucose Restriction on Babies 42:31 Research Findings on Blood Glucose Thresholds 44:24 Managing Gestational Diabetes with Diet and Exercise 48:50 Alternative Testing Methods for Gestational Diabetes 01:00:06 Understanding HbA1c and Its Limitations 01:08:04 Postpartum Care for Babies of Gestational Diabetic Mothers 01:12:42 Connecting with Melissa and Doula Training Opportunities 01:14:40 Conclusion and Final Thoughts   Guest Bio: Melissa Chappell, LDEM, CPM is a midwife and owner of two birth centers in Utah, and as well is a doula trainer of over 22 years. She is passionate about women's health from a holistic and nourishing perspective, and advocates for women's wellness in all areas of their lives. She has worked with midwives and birthing women all over the world, including in Haiti, Ethiopia, and Kenya, and loves seeing how women's lives improve with access to safe and effective midwifery care. She is the mother of 4 children and 4 grandchildren that she adores. In between catching babies, Melissa loves to explore as much of the world as she can – from international travel to exploring the mountains in her backyard. INSTAGRAM: Connect with HeHe on IG  Connect with Melissa on IG    BIRTH EDUCATION: Join The Birth Lounge here for judgment-free childbirth education that prepares you for an informed birth and how to confidently navigate hospital policy to have a trauma-free labor experience!   Download The Birth Lounge App for birth & postpartum prep delivered straight to your phone!   LINKS MENTIONED: Lily Nichols: Real Food in Pregnancy    Use code HeHe to sign up for doula education and 25% off the Birth Education Library with Melissa at https://www.melissachappell.com/   RESEARCH: https://www.cochranelibrary.com/web/cochrane/content?templateType=full&urlTitle=/cdsr/doi/10.1002/14651858.CD012394.pub3&doi=10.1002/14651858.CD012394.pub3&type=cdsr&contentLanguage=   https://bmcpregnancychildbirth.biomedcentral.com/articles/10.1186/s12884-023-05779-z   Gestational diabetes and the risk of late stillbirth: a case-control study from England, UK - PubMed   https://www.cochranelibrary.com/web/cochrane/content?templateType=full&urlTitle=/cdsr/doi/10.1002/14651858.CD011624.pub3&doi=10.1002/14651858.CD011624.pub3&type=cdsr&contentLanguage=   Big Babies: the risk of care provider fear | Dr Rachel Reed  

Episode 193: Gestational Diabetes IntroJesica Mendoza (OMSIII) describes the pathophysiology of gestational diabetes and the right timing and method of screening for it. Dr. Arreaza adds insight into the need for culturally-appropriate foods, such as vegetables in Mexican cuisine.    Written by Jesica Mendoza, OMSIII, Western University of Health Sciences, College of Osteopathic Medicine of the Pacific. Editing by Hector Arreaza, MD.You are listening to Rio Bravo qWeek Podcast, your weekly dose of knowledge brought to you by the Rio Bravo Family Medicine Residency Program from Bakersfield, California, a UCLA-affiliated program sponsored by Clinica Sierra Vista, Let Us Be Your Healthcare Home. This podcast was created for educational purposes only. Visit your primary care provider for additional medical advice.DefinitionGestational diabetes mellitus (GDM) is a condition that occurs to previously non-diabetic pregnant women, caused by glucose intolerance at around the 24th week of gestation. PathophysiologyGDM arises due to an underlying pancreatic beta cell dysfunction in the mother which leads to a decrease in the amount of insulin produced and thus leads to higher blood sugar levels during pregnancy. The placenta of the fetus will produce hPL (human placental lactogen) to ensure a steady supply of sugars to the fetus, creating an anti-insulin effect. However, hPL readily crosses the placental barrier causing the mothers insulin requirement to increase, when the mother's pancreas cannot increase production of insulin to that level needed to counter the effect of hPL they become diabetic, and this leads to gestational diabetes. So, basically the placenta is asking for more glucose for the baby and the mother's pancreas struggles to keep the glucose level within normal limits in the body of the mother. If left untreated, high levels of glucose in the mother can cause glucotoxicity in the mother.“Glucotoxicity” refers to the toxic effect of glucose. Glucose is the main fuel for cell functions, but when it is high in the bloodstream, it causes toxicity to organs. Prevalence of GDM.The CDC reports mean prevenance of GDM is 6.9%. In U.S. mothers the prevenance increased from 6.0% in 2016 to 8.3% in 2021. Many different factors have played a role in increasing gestational diabetes in American mothers, some of those being the ongoing obesity epidemic with excess body weight being a known risk factor for insulin resistance. Another being advanced maternal age (AMA) as more American women have children later in life their body becomes less sensitive to insulin and requires a higher insulin output on top of the insulin that is required for the fetus. The “American diet” is also something that has a big effect in diabetes development. With the increase of high-carb foods that are readily available, the diet of Americans has declined and is affecting the metabolic health of mothers as they carry and deliver their children. Despite ongoing awareness of GDM, 6% to 9% of pregnant women in the United States are diagnosed with gestational diabetes, and the prevalence continues to increase worldwide. It is estimated that in 2017 18.4 million pregnancies were affected by GDM in the world, which then continued to increase to 1 in 6 births to women with GDM in 2019. It was also found that women living in low-income communities were disproportionately affected due to limited healthcare access. Additionally, women with GDM had a 1.4-fold increase in likelihood of undergoing a c-section, with 15% increase in risk of requiring blood transfusion. Screening for GDMGestational diabetes is screened between the 24th to 28th week of gestation in all women without known pregestational diabetes. In women who have high-risk for GDM the screening occurs during the first trimester, these women usually have at least one of the following: BMI > 30, prior history of GDM, known impaired glucose metabolism, and/or a strong family history of diabetes. The screening during the first trimester is to detect “pregestational diabetes” because we have to keep a good glycemic control to improve outcomes of pregnancy. So, if it's positive, you start treatment immediately. If these women are found to have a normal glucose, they repeat the testing again as done normally, at 24-28 weeks of gestation. How do we screen?The screening itself consists of two types of approaches. The two-step approach includes a 50-gram oral glucose tolerance test (OGTT), where blood glucose is measured in an hour and if it is below 140 they are considered to not have GDM, however if the reading is greater than 140 they must then do a 3-hour, 100g oral glucose tolerance test. The 3-hour OGTT includes measuring the blood sugars at Fasting which should be less than 95, at 1 hour at less than 180, at 2 hours at less than 155, and at 3 hours at less than 140. If 2 or more of these values exceed the threshold the patient is diagnosed with gestational diabetes mellitus. The one-step approach includes 75g after an overnight fast. Blood glucose is measured while fasting which should be less than 92, at 1 hour less than 180 and at 2 hours less than 153. If any one of these values is exceeded, the patient is diagnosed with GDM.If the mother is found to be GDM positive during pregnancy she will also need continued screening post-partum to monitor for any development of overt diabetes. The testing is usually 75g 2-hour OGTT at 6-12 weeks postpartum. If this testing is normal, then they are tested using HbA1c every 3 years. If the post-partum testing shows pre-diabetes, annual testing is recommended using HbA1c measurements. Maternal complications Women with GDM are at an increased risk for future cardiovascular disease, T2DM, and chronic kidney disease. GDM is also associated with increased likelihood of developing pre-eclampsia following delivery. Pre-eclampsia is a complication seen in pregnancy characterized by high blood pressure, proteinuria, vision changes, and liver involvement (high LFTs). Pre-eclampsia can then progress to eclampsia or HELLP syndrome, both of which can include end organ damage. Additionally, she can develop polyhydramnios which leads to overstretching of the uterus and can induce pre-term labor, placental abruption, and or uterine atony, all of which additionally put the mother at increased risk for c-section. All of these maternal complications that stem from GDM lead to complications and extended hospitalization. Child's complications Although there is an increased set of risks for the mother, the neonate can also develop a variety of risks due to the increased glucose while in utero. While the fetus is growing, the placenta is the source of nutrition for the fetus. As the levels of glucose in the mother increase so does the amount of glucose filtered through the placenta and into the fetal circulation. Over time the glucose leads to oxidative stress and inflammation with activation of TGF-b which leads to fibroblast activation and fibrosis of the placenta. This fibrosis decreases the nutrient and oxygen exchange for the fetus. As the fetus attempts to grow in this restrictive environment its development is affected. The fetus can develop IUGR (intrauterine growth restriction) leading to a small for gestation age newborn which can then lead to another set of complications. The low oxygen environment can lead to increased EPO production and polycythemia at birth which can then lead to increased clotting that can travel to the newborn brain. Newborns can also be born with fetal acidosis due to the anerobic metabolism and lactic acid buildup in fetal tissues which can cause fetal encephalopathy leading to cerebral palsy and developmental delay. And the most severe of newborn complications to gestational diabetes can lead to fetal demise. Furthermore, the increase of glucose can also lead to macrosomia in the infant which can often lead to a traumatic delivery and delivery complications such as shoulder dystocia and brachial plexus injury. Brachial plexus injury sometimes resolves without sequela, but other times can lead to permanent weakness or paralysis of the affected arm. The baby can be born too small or too big.Additionally, once the fetus is born the cutting of the umbilical cord leads to a rapid deceleration in blood glucose in the fetal circulation and hypoglycemic episodes can occur, that often lead to NICU admission. The insulin that is created by the fetus in utero to accommodate the large quantities of glucose is known to affect lung maturation as well. The insulin produced inhibits surfactant production in the fetus. Upon birth some of the newborns also have to be placed on PEEP for ventilation and some children require treatment with surfactant to prevent alveolar collapse and/or progression to NRDS created by the low surfactant levels. Additionally, neonates who are macrosomic, which is usually seen in GDM mothers, are larger and stronger and when put on PEEP to help increase ventilation the newborn's stronger respiratory effort can lead to higher pulmonary pressures and barotrauma such as neonatal pneumothorax.Long term complications to the child of a mother with GDM also occur. As the child grows, they are also at an increased risk for developing early onset obesity because of the increased adipose storage triggered by the increase in insulin in response to the high glucose in utero. This then can lead to a higher chance of developing type 2 diabetes mellitus in the child. With diabetes, also comes an increase in cardiovascular risk as the child ages and becomes an adult. The effects of GDM go beyond the fetal life but continue through adulthood.What can be done?Gestational Diabetes Mellitus has many severe and lifelong consequences for both the mother and the child and prevention of GDM would help enhance the quality of life of both. Many of the ways to prevent GDM complications include patient education and dietary modifications with a diet rich in whole grains, fruits, vegetables and lean proteins. Benefits of some vegetables in the Mexican cuisine that may be beneficial: Nopales, Chayote, and Jicama. Those are good alternatives for highly processed carbs.Mothers are usually offered nutritional counseling to help them develop a tailored eating plan. This and 30 minutes of moderate exercise daily is recommended to increase insulin sensitivity and lower the post-prandial glucose levels. If within 2 weeks of implementing lifestyle changes alone the glucose measurements remain high, then medications like insulin can be put onboard to manage the GDM. If they require insulin, I think it is time to refer to a higher level of care, if available, high risk OB clinic.Conclusion: Now we conclude episode number ###, “[TITLE].” [summary here]. _____________________References:Eades CE, Burrows KA, Andreeva R, Stansfield DR, Evans JM. Prevalence of gestational diabetes in the United States and Canada: a systematic review and meta-analysis. BMC Pregnancy Childbirth. 2024 Mar 15;24(1):204. doi: 10.1186/s12884-024-06378-2. PMID: 38491497; PMCID: PMC10941381. https://pubmed.ncbi.nlm.nih.gov/38491497/QuickStats: Percentage of Mothers with Gestational Diabetes,* by Maternal Age — National Vital Statistics System, United States, 2016 and 2021. Weekly / January 6, 2023 / 72(1);16. https://www.cdc.gov/mmwr/volumes/72/wr/mm7201a4.htm?utmAkinyemi OA, Weldeslase TA, Odusanya E, Akueme NT, Omokhodion OV, Fasokun ME, Makanjuola D, Fakorede M, Ogundipe T. Profiles and Outcomes of Women with Gestational Diabetes Mellitus in the United States. Cureus. 2023 Jul 4;15(7):e41360. doi: 10.7759/cureus.41360. PMID: 37546039; PMCID: PMC10399637. https://pmc.ncbi.nlm.nih.gov/articles/PMC10399637/?utmPerlman, J. M. (2006). Summary proceedings from the neurology group on hypoxic-ischemic encephalopathy. Pediatrics, 117(3), S28–S33.DOI: 10.1542/peds.2005-0620C.Low, J. A. (1997). Intrapartum fetal asphyxia: definition, diagnosis, and classification. American Journal of Obstetrics and Gynecology, 176(5), 957–959.DOI: 10.1016/S0002-9378(97)70609-0.Hallman, M., Gluck, L., & Liggins, G. (1985). Role of insulin in delaying surfactant production in the fetal lung. Journal of Pediatrics, 106(5), 786–790.DOI: 10.1016/S0022-3476(85)80227-0.Sweet, D. G., Carnielli, V., Greisen, G., et al. (2019). European Consensus Guidelines on the Management of Respiratory Distress Syndrome – 2019 Update. Neonatology, 115(4), 432–450.DOI: 10.1159/000499361.Raju, T. N. K., et al. (1999). Respiratory distress in term infants: when to suspect surfactant deficiency. Pediatrics, 103(5), 903–909.DOI: 10.1542/peds.103.5.903.Burns, C. M., Rutherford, M. A., Boardman, J. P., & Cowan, F. M. (2008). Patterns of cerebral injury and neurodevelopmental outcomes after symptomatic neonatal hypoglycemia. Pediatrics, 122(1), 65–74.DOI: 10.1542/peds.2007-2822.Dabelea, D., et al. (2000). Long-term impact of maternal diabetes on obesity in childhood. Diabetes Care, 23(10), 1534–1540.DOI: 10.2337/diacare.23.10.1534.Dashe, J. S., et al. (2002). "Hydramnios: Etiology and outcome." Obstetrics & Gynecology, 100(5 Pt 1), 957–962.DOI: 10.1016/S0029-7844(02)02279-6.Long-term cost-effectiveness of implementing a lifestyle intervention during pregnancy to prevent gestational diabetes mellitus: a decision-analytic modelling study. Diabetologia.American College of Obstetricians and Gynecologists. (2018). Practice Bulletin No. 190: Gestational Diabetes Mellitus. Obstetrics & Gynecology, 131(2), e49–e64. https://doi.org/10.1097/AOG.0000000000002501Theme song, Works All The Time by Dominik Schwarzer, YouTube ID: CUBDNERZU8HXUHBS, purchased from https://www.premiumbeat.com/. 

In a heartfelt conversation, Taime Downe opens up about coping with the loss of his fiancée Kim and how it reinforced his commitment to sobriety. Despite tour challenges, including an early bus breakdown, he's finding strength through music and his band family. Downe also discusses Faster Pussycat's latest single 'Motorbike,' their approach to releasing new music, and their extensive upcoming tour with The Supersuckers.Plus, former Guns N' Roses and Sixx:AM guitarist DJ Ashba discusses his innovative venture into GDM, blending rock guitar with dance music. He shares insights about performing at Vegas clubs, his creative process, and the challenges of bridging two distinct musical worlds. Ashba also discusses the status of Sixx:AM, his memorable moments with GNR, and how his successful fabrication business allows him creative freedom to explore new musical territories. Catch Eddie Trunk every M-F from 3:00-5:00pm ET on Trunk Nation on SiriusXM Faction Talk Channel 103.And don't forget to follow Eddie on Twitter and Instagram!Follow the link to get your free 3-month trial of SiriusXM: http://siriusxm.com/eddietrunk Find all episodes of Trunk Nation: https://siriusxm.com/trunknation

It's In the News.. a look at the top headlines and stories in the diabetes community. This week's top stories: Learning more about the FDA letter sent to Dexcom, news from ATTD including a bihormonal pump from a Dutch company, time in tight range update, more studies about using insulin and GLP-1 medications, eating chili to prevent gestational diabetes (really!) and more..  Find out more about Moms' Night Out  Please visit our Sponsors & Partners - they help make the show possible! Learn more about Gvoke Glucagon Gvoke HypoPen® (glucagon injection): Glucagon Injection For Very Low Blood Sugar (gvokeglucagon.com) Omnipod - Simplify Life Learn about Dexcom   Check out VIVI Cap to protect your insulin from extreme temperatures The best way to keep up with Stacey and the show is by signing up for our weekly newsletter: Sign up for our newsletter here Here's where to find us: Facebook (Group) Facebook (Page) Instagram Twitter Check out Stacey's books! Learn more about everything at our home page www.diabetes-connections.com  Reach out with questions or comments: info@diabetes-connections.com Episode transcription with links: Hello and welcome to Diabetes Connections In the News! I'm Stacey Simms and every other Friday I bring you a short episode with the top diabetes stories and headlines happening now. XX Our top story this week: Dexcom Dive Brief: A warning letter posted Tuesday by the Food and Drug Administration revealed quality control issues with Dexcom's continuous glucose monitors. The FDA raised concerns with a design change to a component used in the resistance layer of Dexcom's sensors. The sensors with the new component were less accurate than those with the original component, according to the warning letter. Dexcom has ceased distribution of G7 sensors with the component, but the company's response did not address affected G6 sensors. J.P. Morgan analyst Robbie Marcus wrote in a research note Tuesday that the letter concerns a chemical compound that the sensor wire is dipped in. Dexcom began producing the compound internally to add redundancy to its supply chain.   Dive Insight: Dexcom Chief Operating Officer Jake Leach said in an interview with MedTech Dive last week that the company does not expect the warning letter to affect future product approvals, including a 15-day version of its G7 CGM, and there's no need yet to recall products. Dexcom has submitted the device to the FDA and anticipates a launch in the second half of the year.   Marcus, after speaking to company leadership and a quality control expert, wrote that many of the issues outlined in the letter could be addressed quickly. He added that the warning letter could explain minor delays in approval to the 15-day sensor, but Dexcom is still within the 90-day window for a 510(k) submission.   “While there's always a risk this could impede future product approvals,” Marcus wrote, “we do not expect this to materially delay the 15 day G7 sensor approval.”   The warning letter followed an FDA inspection last year of Dexcom's facilities in San Diego and Mesa, Arizona. Marcus wrote that after the FDA requested additional information and a separate 510(k), Dexcom stopped in-sourcing the compound and reverted back to the external supplier.   Dexcom's devices were misbranded because the company did not submit a premarket notification to the FDA before making major changes to the sensors, according to the warning letter. The sensors with the changed coating “cause higher risks for users who rely on the sensors to dose insulin or make other diabetes treatment decisions,” the letter said.     The FDA raised other concerns in the warning letter, including procedures to monitor the glucose and acetaminophen concentrations used in testing of the G6 and G7 CGMs. The FDA also cited problems with Dexcom's handling last year of a deficiency in its G6 sensors with dissolved oxygen content values, a key input for measuring blood glucose levels. https://www.medtechdive.com/news/dexcom-warning-letter-cgm-coating-change/743597/ XX Lots of studies and info out of the recent ATTD conference. One highlight that has been sort of under the radar: a Dutch company has been using a Bihormonal fully closed-loop system for the treatment of type 1 diabetes in the real world. This is a company called Inreda (in-RAY-duh). The Inreda AP® is an automatic system (closed loop) and independently regulates the blood glucose level by administering insulin and glucagon. The AP5 is certified in Europe and is being used in multiple studies and projects. The AP®6 is currently under development. https://www.inredadiabetic.nl/en/discover-the-ap/ https://pubmed.ncbi.nlm.nih.gov/38443309/ XX Let's talk about time in tight range. If you follow me and diabetes connections on social, you likely saw a video I made about this – it blew up last week. If not.. time in range has been a metric for a short while now.. in 2019 there was a consensus report advising a goal of 70% of time in the 70-180 mg/dL range for most people with type 1 diabetes (T1D) and type 2 diabetes (T2D), with modifications for certain subgroups. Recently we've been hearing more about 70-140 mg/dL — for longer periods as “time in tight range (TiTR).” At ATTD there was more talk about calling that range TING, or “time in normal glycemia.     There's a great writeup that I'll link up from the great Miriam Tucker on Medscape about a debate that happened at ATTD. On March 22, 2025, two endocrinologists debated this question at the Advanced Technologies & Treatments for Diabetes (ATTD) 2025. Anders L. Carlson, MD, medical director of the International Diabetes Center (IDC), Minneapolis, took the positive side, while Jeremy Pettus, MD, assistant professor of medicine at the University of California San Diego, who lives with T1D himself, argued that it's too soon.   https://www.medscape.com/viewarticle/should-time-tight-range-be-primary-diabetes-goal-2025a100073q?form=fpf   XX Sequel Med Tech announces its twist pump will be firs paired with Abbott's FreeStyle Libre 3 Plus. The twist has FDA approval for ages 6 and up and is set to begin its commercial launch by the end of June. The pump—designed by inventor Dean Kamen's Deka Research & Development—also incorporates the FDA-cleared Tidepool Loop software program, to record CGM blood sugar readings, make predictions based on trends and adjust its background insulin levels accordingly. https://www.fiercebiotech.com/medtech/sequel-med-tech-connects-twiist-insulin-pump-abbotts-cgm-ahead-market-debut XX Dexcom's longer-lasting CGM sensor looks promising, based on study results presented at the conference. The trial showed that the new 15-day G7 system is slightly more accurate than the current G7. The accuracy of CGM can be measured using MARD (mean absolute relative difference), which shows the average amount a CGM sensor varies from your actual glucose levels (a lower number is better).  The 15-day G7 has a MARD value of 8.0%, about the same as the Abbott Freestyle Libre 3. The Dexcom G7 15 Day is awaiting FDA approval and is not yet available in the U.S.   XX Little bit of news from Modular Medical.. they plan to submit their patch pump to the FDA late summer or fall of this year. The MODD1 product, a 90-day patch pump, features new microfluidics technology to allow for the low-cost pumping of insulin. Its new intuitive design makes the product simple to use and easier to prescribe. It has a reservoir size of 300 units/3mL. Users can monitor the pump activity with their cell phone and do not require an external controller. The pump uses a provided, single-use, disposable battery. Modular Medical picked up FDA clearance for MODD1 in September. The company also raised $8 million to end 2024. Its founder, Paul DiPerna, previously founded leading insulin pump maker Tandem Diabetes Care. DiPerna invented and designed Tandem's t:slim pump. By developing its patented insulin delivery technologies, the company hopes to improve access to glycemic control. Its founder, Paul DiPerna, previously founded leading insulin pump maker Tandem Diabetes Care. DiPerna invented and designed Tandem's t:slim pump. https://www.drugdeliverybusiness.com/modular-medical-announces-12m-private-placement/ XX More from attd – type 2 news? https://www.drugdeliverybusiness.com/biggest-diabetes-tech-news-attd-2025/ XX Another study that says people with type 1 who use a GLP-1 medication get better outcomes. In this study, those who use GLP-1 with insulin are 55% less likely to have a hyperglycemia-related ED visit, 26% less likely to have an amputation-related visit, and 29% less likely to have a diabetic ketoacidosis (DKA)-related ED visit in the following year compared to those on insulin alone. Although they are not approved for T1D, some patients may receive them off-label or for weight control. Pretty big study for an off label drug: compared 7,010 adult patients with T1D who were prescribed GLP-1s and insulin to 304,422 adult patients with T1D who were on insulin alone.  It is important to note that the rates of new diabetic complications in one year for both groups were around 1%, indicating that these are uncommon outcomes regardless of medication use. https://www.epicresearch.org/articles/some-diabetic-complications-less-likely-among-type-1-diabetics-on-glp-1s   XX Early research here but exposure to antibiotics during a key developmental window in infancy may stunt the growth of insulin-producing cells in the pancreas and boost risk of diabetes later in life The study, is published this month in the journal Science, it's a study in mice. These researchers are working off the idea that when while identical twins share DNA that predisposes them to Type 1 diabetes, only one twin usually gets the disease. She explained that human babies are born with a small amount of pancreatic “beta cells,” the only cells in the body that produce insulin.   But some time in a baby's first year, a once-in-a-lifetime surge in beta cell growth occurs.   “If, for whatever reason, we don't undergo this event of expansion and proliferation, that can be a cause of diabetes,” Hill said.   They found that when they gave broad-spectrum antibiotics to mice during a specific window (the human equivalent of about 7 to 12 months of life), the mice developed fewer insulin producing cells, higher blood sugar levels, lower insulin levels and generally worse metabolic function in adulthood.   in other experiments, the scientists gave specific microbes to mice, and found that several they increased their production of beta cells and boosted insulin levels in the blood. When male mice that were genetically predisposed to Type 1 diabetes were colonized with the fungus in infancy, they developed diabetes less than 15% of the time. Males that didn't receive the fungus got diabetes 90% of the time. Even more promising, when researchers gave the fungus to adult mice whose insulin-producing cells had been killed off, those cells regenerated. Hill stresses that she is not “anti-antibiotics.” But she does imagine a day when doctors could give microbe-based drugs or supplements alongside antibiotics to replace the metabolism-supporting bugs they inadvertently kill.   .   “Historically we have interpreted germs as something we want to avoid, but we probably have way more beneficial microbes than pathogens,” she said. “By harnessing their power, we can do a lot to benefit human health.”     https://www.eurekalert.org/news-releases/1078112 XX Future watch for something called BeaGL - created by researchers at the University of California Davis and UC Davis Health who were inspired by their own personal experiences with managing T1D.   BeaGL is designed to work with CGMs and has security-focused machine learning algorithms to make predictive alerts about anticipated glucose changes, which are sent to a device. In this case, a smartwatch. The end goal is for BeaGL to be completely automated to reduce the cognitive load on the patient, particularly for teens. It's still in research phase but six student with T1D have been using it for almost a year.     https://health.ucdavis.edu/news/headlines/with-ai-a-new-metabolic-watchdog-takes-diabetes-care-from-burden-to-balance/2025/02 XX Investigators are searching for a way forward after two long-term diabetes programs were terminated following the cancellation of their National Institutes of Health (NIH) funding, the result of federal allegations that study coordinator Columbia University had inappropriately handled antisemitism on campus. The programs include the three-decades-old Diabetes Prevention Program (DPP) and its offshoot, the Diabetes Prevention Program Outcomes Study (DPPOS). “We are reeling,” said David Nathan, MD, a previous chair of both the DPP and the DPPOS and an original leader of the landmark Diabetes Control and Complications Trial. Nathan is also founder of the Massachusetts General Hospital Diabetes Center in Boston, one of the 30 DPPOS sites in 21 states. On March 7, the Trump administration cancelled $400 million in awards to Columbia University from various federal agencies. While Columbia University agreed on March 21 to changes in policies and procedures to respond to the Trump administration's charges, in the hopes that the funding would be restored, DPPOS Principal Investigator Jose Luchsinger, MD, told Medscape Medical News that as of press time, the study was still cancelled. https://www.medscape.com/viewarticle/diabetes-prevention-program-cancellation-colossal-waste-2025a100076h XX XX Type 2 diabetes may quietly alter the brain in ways that mimic early Alzheimer's. This was only an animal study – but researchers say the high comorbidity of type 2 diabetes (T2D) with psychiatric or neurodegenerative disorders points to a need for understanding what links these diseases.   https://scitechdaily.com/how-diabetes-quietly-rewires-the-brains-reward-and-memory-system/ XX Eating chili once a month when you're pregnant seems to lower the risk of developing gestational diabetes. This is a real study! While chili showed a link to lower gestational diabetes risk, dried beans and bean soup had no significant effect, even among women who ate them more frequently. Some studies suggest that diets high in beans and legumes, including the Mediterranean diet, reduce GDM risk. While studies link beans to lower diabetes risk, their specific impact on GDM remains unclear. This study analyzed data from 1,397 U.S. pregnant women who participated in the Infant Feeding Practices Study II, conducted between 2005 and 2007. Chili consumption varied significantly by race, education, household size, income, supplemental nutrition status, and region. Non-Hispanic Black mothers consumed the most (0.33 cups/week), while those with higher income and education levels consumed less. Regional differences also influenced chili intake. One possible mechanism for chili's effect is capsaicin, a bioactive compound found in chili peppers, which has been linked to metabolic benefits in other studies. However, further research is needed to confirm this potential role in GDM prevention. Dried bean and bean soup consumption had no clear association with GDM. The study highlights limitations due to self-reported dietary data and the need for more detailed dietary measures. https://www.news-medical.net/news/20250317/Could-a-little-spice-in-your-diet-prevent-gestational-diabetes.aspx XX

One in five women in the U.S. have a BMI of 30 or more at the START of pregnancy. Around 1 in 5 women gain more than 40 pounds during pregnancy, which is more than any woman should gain. Only about one-third of women gain the recommended amount of weight during pregnancy. Gaining too much weight during pregnancy can increase the risk of HDP, GDM, fetal macrosomia, and can cause complications of birth, such as shoulder dystocia or preterm birth. Excessive weight gain during pregnancy can also increase the likelihood of postpartum weight retention. But what about stillbirth risk? Does excessive maternal weight gain during pregnancy increase still birth risk? The ACOG recommends antepartum fetal surveillance based on pre-pregnancy BMI. Why is maternal weight during pregnancy not an indication for an antepartum fetal surveillance? The data may surprise you! Listen in for details.

Understanding Gestational Diabetes – Risks, Complications & Treatment In this episode of MamaDoc BabyDoc, we dive into gestational diabetes—a condition that affects nearly 10% of pregnancies. Join our OB/Gyn and pediatrician duo as we break down the risk factors, potential complications for both mom and baby, and the best strategies for managing blood sugar during pregnancy. We'll also discuss how gestational diabetes can impact long-term health and what steps you can take to ensure a healthy pregnancy and delivery. Whether you're currently expecting, planning for pregnancy, or just curious about the topic, this episode is packed with essential information every parent should know!

(We were made aware that this original posting had the last section DROPPED accidentally)...here is the full episode! Ahhh...TECHNOLOGY! *This is why AI will likely replace our production team...Just kidding production team, just kidding).Episode Details:Well, we typically focus on ONE or maybe TWO publications to highlight and review. However, in this episode, which we have decided to call, “Survey said…!”, we will go through some common and REAL WORLD “mental battles”regarding what is and what is not part of a diagnostic criteria. These are every day OBGYN things that we KNOW, but when asked to define them…we can easily get ourselves confused. We are going to clear these up…Game Show style!  First, when only one abnormal value is found in the two-step, 100-gram GTT,  it is called borderline GDM, or impaired glucose tolerance. But what is it called when there is an abnormal (failed) 1-Hour 50 gram, but completely normal 3-Hr 100-gram GTT? Is this also called “impaired glucose tolerance”? We….the Survey Said…! (Yep, we'll get to that). Secondly, does the criteria for Preeclampsia with Severe Criteria include platelets of 100,000 or not? The Survey Said…! (Yep, we'll cover that). We will also review the numbers for MVP oligo, for a “normal” postmenopausal ES, and MORE! Listen in for details!

Well, we typically focus on ONE or maybe TWO publications to highlight and review. However, in this episode, which we have decided to call, “Survey said…!”, we will go through some common and REAL WORLD “mental battles” regarding what is and what is not part of a diagnostic criteria. These are every day OBGYN things that we KNOW, but when asked to define them…we can easily get ourselves confused. We are going to clear these up…Game Show style! First, when only one abnormal value is found in the two-step, 100-gram GTT, it is called borderline GDM, or impaired glucose tolerance. But what is it called when there is an abnormal (failed) 1-Hour 50 gram, but completely normal 3-Hr 100-gram GTT? Is this also called “impaired glucose tolerance”? We….the Survey Said…! (Yep, we'll get to that). Secondly, does the criteria for Preeclampsia with Severe Criteria include platelets of 100,000 or not? The Survey Said…! (Yep, we'll cover that). We will also review the numbers for MVP oligo, for a “normal” postmenopausal ES, and MORE! Listen in for details!

In this episode, we will cover 2 topics: the first is brand new in print (01/06/2025 ), and the second is just weird. In the “new” portion we'll summarize a new randomized study published in JAMA Network dealing with gestational diabetes. Should we add glyburide to metformin for GDM control? Listen in for details. In the second portion, we'll focus on unilateral ovarian absence not related to previous removal. Yep! This is why it's very important to check the adnexa at “routine” C-section or “routine” gynecological surgery. It is possible to be missing an ovary…and its weird! Listen in for details!

In this episode, we sit down with Tom Hopcroft, the British entrepreneur and founder of Guiris de Mierda, a vibrant community bringing expats and locals together in Madrid. Tom shares his journey from the UK's corporate grind to creating a network that fosters authentic connections, unforgettable events, and a sense of belonging.We dive into what it means to be a "guiri," the challenges of building a community-based business abroad, and how Tom's adventures—like walking the Camino de Santiago and skateboarding across Spain—have shaped his approach to life and work.From navigating cultural differences to crafting viral content, building brand partnerships, and scaling GDM to other cities, this episode is packed with insights for anyone curious about expat life, entrepreneurship, or the art of growing a mission-driven brand.Tune in for a lively conversation about community, culture, and making it work abroad------Timestamps:02:16 What is a Guiri?05:57 GDM Events, Investing in Spain, More than "Siesta y Fiesta"09:55 Journey from Merch to Content to the 1st Meet-up17:07 Why the Name "Guiris de Mierda"?17:59 When Did You Realize This Was a Business?21:24 Golden Guiris and the Subscription Model25:31 Learning From Other Communities27:43 Viral Content to Booked Events Flywheel28:32 Finding Your Niche31:39 Skateboarding Across Spain33:42 Content Strategy (Personal v Business Brand)37:40 Hosting and Organizing Live Events40:03 Getting Sponsorships and Partnerships44:51 GDM Expanding to Barcelona (1st Event)46:00 Content Creation Workshop49:42 Questions From the Audience56:18 Rapid Fire Questions------Follow Tom:https://www.instagram.com/tomcharliedesignFollow Guiris de Mierda:https://www.instagram.com/guirisdemierdahttps://linktr.ee/tomcharliedesignWatch Tom's Backstory:https://www.youtube.com/watch?v=1BOfWd4lts4

In this episode, Trista explores the complexities of managing pregnancy with PCOS, focusing on dietary considerations, the safety of medications and supplements, and the importance of mental health during the postpartum period. She emphasizes the need for careful monitoring of blood sugar levels, the role of insulin and metformin, and the significance of support systems for new parents. You'll learn: Why managing glycemic load is crucial for pregnant individuals with PCOS Safe and effective treatments for gestational diabetes How postpartum mental health is a significant concern for new parents Navigating potential chest feeding challenges for those with PCOS Episode Links: How to Manage Gestational Diabetes with Diet and Lifestyle 1-on-1 Nutrition Coaching References: Choudhury, A. A., & Rajeswari, V. D. (2022). Polycystic ovary syndrome (PCOS) increases the risk of subsequent gestational diabetes mellitus (GDM): A novel therapeutic perspective. Life Sciences (1973), 310, 121069–121069. https://doi.org/10.1016/j.lfs.2022.121069 Diabetes Canada. (2024). Gestational diabetes. https://www.diabetes.ca/about-diabetes/gestational Facchinetti, F., Cavalli, P., Copp, A. J., D'Anna, R., Kandaraki, E., Greene, N. D. E., & Unfer, V. (2020). An update on the use of inositols in preventing gestational diabetes mellitus (GDM) and neural tube defects (NTDs). Expert Opinion on Drug Metabolism & Toxicology, 16(12), 1187–1198. https://doi.org/10.1080/17425255.2020.1828344 Ibrahim, I., Bashir, M., Singh, P., Al Khodor, S., & Abdullahi, H. (2022). The Impact of Nutritional Supplementation During Pregnancy on the Incidence of Gestational Diabetes and Glycaemia Control. Frontiers in Nutrition (Lausanne), 9, 867099–867099. https://doi.org/10.3389/fnut.2022.867099 Jorquera, G., Echiburú, B., Crisosto, N., Sotomayor-Zárate, R., Maliqueo, M., & Cruz, G. (2020). Metformin during Pregnancy: Effects on Offspring Development and Metabolic Function. Frontiers in Pharmacology, 11, 653–653. https://doi.org/10.3389/fphar.2020.00653 Koric, A., Singh, B., VanDerslice, J. A., Stanford, J. B., Rogers, C. R., Egan, D. T., Agyemang, D. O., & Schliep, K. (2021). Polycystic ovary syndrome and postpartum depression symptoms: a population-based cohort study. American Journal of Obstetrics and Gynecology, 224(6), 591.e1-591.e12. https://doi.org/10.1016/j.ajog.2020.12.1215 Ryssdal, M., Vanky, E., Stokkeland, L. M. T., Jarmund, A. H., Steinkjer, B., Løvvik, T. S., Madssen, T. S., Iversen, A.-C., & Giskeødegård, G. F. (2023). Immunomodulatory Effects of Metformin Treatment in Pregnant Women With PCOS. The Journal of Clinical Endocrinology and Metabolism, 108(9), e743–e753. https://doi.org/10.1210/clinem/dgad145 Slouha, E., Alvarez, V. C., Gates, K. M., Ankrah, N. M. N., Clunes, L. A., & Kollias, T. F. (2023). Gestational Diabetes Mellitus in the Setting of Polycystic Ovarian Syndrome: A Systematic Review. Curēus (Palo Alto, CA), 15(12), e50725–e50725. https://doi.org/10.7759/cureus.50725 Vanky, E., Isaksen, H., Haase Moen, M., & Carlsen, S. M. (2008). Breastfeeding in polycystic ovary syndrome. Acta Obstetricia et Gynecologica Scandinavica, 87(5), 531–535. https://doi.org/10.1080/00016340802007676

Det er stenhårdt arbejde at være gravid med diabetes. Hvert år gennemfører ca. 2400 kvinder en graviditet, samtidig med at de har diabetes. Nogle af dem har type 1 eller type 2-diabetes med sig ind i graviditeten, og andre udvikler en særlig form for diabetes i graviditeten, også kaldet graviditetsdiabetes (GDM). Sammen med professor og overlæge Elisabeth R. Mathiesen, der til daglig er tillnyttet Center for Gravide med Diabetes på Rigshospitalet, bliver vi her klogere på, hvordan det er være gravid med diabetes. Hvad er det for komplikationer og risici, man er opmærksom på? Hvilket graviditetsforløb kan man forvente? Hvad er behandlingsforløbet, og hvad skal man selv være opmærksom på før, under og efter graviditeten?

In November's Green Journal, Drs Amy Valent and Linda Barbour will publish their Clinical Expert Series (CES) on insulin management in GDM and Type 2 DM in pregnancy. This is a FANTASTIC document and is our subject matter in this episode. Here, we will give clinical pearls for insulin initiation in pregnancy based on 3 regimens (NPH/Reg; NPH/RAAs; Basal-Bolus) and their initiation in an easy to follow format. Congratulations to Drs Valent and Barbour on a wonderful CES.

On May 22, 2024, we summarized a then soon-to-be-released ACOG CPU on Screening for GDM in Pregnancy and Postpartum. That CPU was officially released July 2024. That update endorsed the possibility of immediate postpartum GTT testing with a 75-gram OGTT. Now, on September 19, 2024, authors from UT Houston have published a systematic review/meta-analysis on this subject. In this episode, we will review what this data is and what it isn't. Listen in for details.

GDM 773 - This episode features a mix from Kaeno. For more, check out www.facebook.com/kaeno.music & @kaeno. Global Dance Mission 773 (Soundcloud & Mixcloud) features Kaeno in the mix! Kaeno is back with an exclusive set designed to ignite your senses. Enjoy an epic, energetic journey with tracks from a variety of top producers! Time to dance… Kaeno is your guide… peace, love, beats… KEEP THE VIBE ALIVE! Tracklist ---- 01. John O'Callaghan – Space & Time (Indecent Noise Lifestream Edit) 02. Paul Webster – Corruption (Original Mix) 03. Tillmann Uhrmacher – The Pride In Your Eyes (Martin Roth Remix) 04. Lustral – I Feel You (John O'callaghan Remix) 05. Igor S – Airforce One (Will Rees Extended Remix) 06. Joyhauser – PULSAR (Original Mix) 07. David Forbes – 12K.MCG (Original Mix) 08. Mark Sherry – Imbecile (Smith & Brown Remix) 09. Mario Piu – Mario Piu – Communication (Indecent Noise Remix) 10. Gigi Dagostino – Bla Bla Bla (Black XS Bootleg) 11. Will Atkinson – Beans (Extended Mix) 12. David Forbes – Randomize 13. Blue Serigala – Come Closer (Will Rees Remix) 14. I.D. 15. Bryan Kearney & Plumb – All Over Again (Karney Dark Dub Extended Mix) 16. Bicep – Glue (Karney Belfast Bootleg) 17. Push – Strange World (2000 Remake) 18. M.I.K.E. Push – Liquid Overdose – Ancient Space (Fred Baker Remix) 19. Mark Sixma, Orjan Nilsen & Push – Urban Shakedown (nilsix Remix) (Extended Mix) 20. Joint Operations Centre – Timelapse (Sean Tyas pres. abstrkt Extended Remix) 21. Calvin Logue – Do What You Want (Robbie Graham Rework) 22. Robbie Seed & Jimmy Chou & Digital Vision – No More Tears (Extended Mix) 23. onTune – Panaceum (Extended Mix) 24. T78 & D72 – Throw This (Extended Mix) 25. Joseph James (IRL) – Darkness (Original Mix) 26. John Meva – Dream & Fly (Extended Mix) 27. Sam Paganini – Rave (Adam Beyer & Layton Giordani Remix) (Connor Woodford Rework) 28. Thomas Schumacher – When I Rock (A.D.H.S. Remix) 29. Inoblivion – When Darkness Falls (Extended Mix) 30. John Askew – Afterburner (Extended Mix) 31. DK8 – Murder Was The Bass (Robbie Van Doe's Tripping Balls Rework) 32. Adam Ellis & Sid Jay – The Last Stylebender (Extended Mix) 33. Derek Ryan – Escape (Extended Mix)

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: How the AI safety technical landscape has changed in the last year, according to some practitioners, published by tlevin on July 26, 2024 on LessWrong. I asked the Constellation Slack channel how the technical AIS landscape has changed since I last spent substantial time in the Bay Area (September 2023), and I figured it would be useful to post this (with the permission of the contributors to either post with or without attribution). Curious if commenters agree or would propose additional changes! This conversation has been lightly edited to preserve anonymity. Me: One reason I wanted to spend a few weeks in Constellation was to sort of absorb-through-osmosis how the technical AI safety landscape has evolved since I last spent substantial time here in September 2023, but it seems more productive to just ask here "how has the technical AIS landscape evolved since September 2023?" and then have conversations armed with that knowledge. The flavor of this question is like, what are the technical directions and strategies people are most excited about, do we understand any major strategic considerations differently, etc -- interested both in your own updates and your perceptions of how the consensus has changed! Zach Stein-Perlman: Control is on the rise Anonymous 1: There are much better "model organisms" of various kinds of misalignment, e.g. the stuff Anthropic has published, some unpublished Redwood work, and many other things Neel Nanda: Sparse Autoencoders are now a really big deal in mech interp and where a lot of the top teams are focused, and I think are very promising, but have yet to conclusively prove themselves at beating baselines in a fair fight on a real world task Neel Nanda: Dangerous capability evals are now a major focus of labs, governments and other researchers, and there's clearer ways that technical work can directly feed into governance (I think this was happening somewhat pre September, but feels much more prominent now) Anonymous 2: Lots of people (particularly at labs/AISIs) are working on adversarial robustness against jailbreaks, in part because of RSP commitments/commercial motivations. I think there's more of this than there was in September. Anonymous 1: Anthropic and GDM are both making IMO very sincere and reasonable efforts to plan for how they'll make safety cases for powerful AI. Anonymous 1: In general, there's substantially more discussion of safety cases Anonymous 2: Since September, a bunch of many-author scalable oversight papers have been published, e.g. this, this, this. I haven't been following this work closely enough to have a sense of what update one should make from this, and I've heard rumors of unsuccessful scalable oversight experiments that never saw the light of day, which further muddies things Anonymous 3: My impression is that infosec flavoured things are a top ~3 priority area a few more people in Constellation than last year (maybe twice as many people as last year??). Building cyberevals and practically securing model weights at frontier labs seem to be the main project areas people are excited about (followed by various kinds of threat modelling and security standards). Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org

Join Dr. Renda Knapp and Dr. Rachel Schultz as they review the routine prenatal tests that are offered in pregnancy.  In this episode they specifically address NIPT, the screen for gestational diabetes and GBS testing and why these tests are important. 

We know research is crucial for making continued advances in diabetes care for all populations. Rachel Stahl-Salzman, MS, RD, CDN, CDCES, and Kerri Knippen, PhD, RDN, LD, BC-ADM, FAND, join us on The Huddle to talk about their latest research projects related to pregnancy in diabetes, some of the outcomes and learnings of each study, and how diabetes care and education specialists can be leaders in this work, even without a research background.View Rachel's research poster diving deeper into this topic here: Annual QIPS Symposium | Weill Department of Medicine (cornell.edu)Learn more about Kerri's project here: https://www.eeds.com/enduring_material.aspx?AIN=005243415&SIN=230144&Display_Portal_Nav=true https://bsmh.zoom.us/rec/play/dNYY9PJAVNjh_wJCglFQuYOU9GYRTC4JYP1xEr3eqd5037qGu1kvWbgs0Mw35SdAhBtm-W66tyZCnDv8.FBeiYIeEMAKXck4F?canPlayFromShare=true&from=share_recording_detail&startTime=1713455116000&componentName=rec-play&originRequestUrl=https%3A%2F%2Fbsmh.zoom.us%2Frec%2Fshare%2F6jgesPiUBq5EIbX8P7K0pzRJ4yKEb-HPxmBMMhqUZxbBBqREek8OvlNR7vh3aQR2.hQwbtfqHOQ8tp3uF%3FstartTime%3D1713455116000Join the poster presentations at #ADCES24 to learn even more about Kerri and Rachel's work! Learn more and register for the conference here: ADCES24 (adcesmeeting.org)Learn more about the ADCES Foundation here: ADCES Foundation Listen to more episodes of The Huddle at adces.org/perspectives/the-huddle-podcast.Learn more about ADCES and the many benefits of membership at adces.org/join.

Kristian is back with the GDM guys to cover more of the upcoming book. They take a deep dive into the rogue changes, wizard changes, and a small contingent of spells that have been shown.

The GDM guys and their 2 guests have a long conversation about the upcoming players handbook for Dungeons and Dragons. While they don't dive to deeply into the classes as of yet there is still an incredible amount of information to cover.

This conversation is all about gestational diabetes with a Registered Dietitian who specializes in the condition -- Leslee Flannery, RD. Listen in to hear: What is gestational diabetes? What causes gestational diabetes / risk factors How can we prevent gestational diabetes Rate of type 2 diabetes AFTER gestational diabetes Risks of gestational diabetes to mom and babyDo you have to be induced early if you have GDM? Best ways to manage blood sugars with OR without gestational DMDiet & other habits Medications - are they needed? What are the options - pros & cons Relationship with food during pregnancy How to handle GDM diagnosis with a history of disordered eatingWhat to do postpartum after a gestational DM diagnosis – screenings, etc. Nutrition tips for postpartum And more! CONNECT WITH LESLEE: Follow her on IG: @gestational.diabetes.nutritionSupport Group / Resources: click here Affiliate Links: Expecting and Empowered – workout app for pregnancy and postpartum – use my code WELLNESSFORTHEWIN to save on your annual app subscription FullWell Fertility – supplements for before, during and after pregnancy (I love their prenatals and fish oil supps and they have lots of others, too!) -- affiliate code WELLNESSFORTHEWINTubby Todd – bath soaps, shampoos, mineral sunscreen, diaper cream and so much more for your little ones! Use my affiliate link for 10% off – click here  Follow me on IG at @wellnessforthewin and @wellnessforthewinpod Check out my blog for healthy recipes & wellness tips! JOIN MY EMAIL LIST HERE! Please be sure to rate, review and subscribe to the podcast!

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Superposition is not "just" neuron polysemanticity, published by Lawrence Chan on April 26, 2024 on The AI Alignment Forum. TL;DR: In this post, I distinguish between two related concepts in neural network interpretability: polysemanticity and superposition. Neuron polysemanticity is the observed phenomena that many neurons seem to fire (have large, positive activations) on multiple unrelated concepts. Superposition is a specific explanation for neuron (or attention head) polysemanticity, where a neural network represents more sparse features than there are neurons (or number of/dimension of attention heads) in near-orthogonal directions. I provide three ways neurons/attention heads can be polysemantic without superposition: non--neuron aligned orthogonal features, non-linear feature representations, and compositional representation without features. I conclude by listing a few reasons why it might be important to distinguish the two concepts. Epistemic status: I wrote this "quickly" in about 12 hours, as otherwise it wouldn't have come out at all. Think of it as a (failed) experiment in writing brief and unpolished research notes, along the lines of GDM or Anthropic Interp Updates. Introduction Meaningfully interpreting neural networks involves decomposing them into smaller interpretable components. For example, we might hope to look at each neuron or attention head, explain what that component is doing, and then compose our understanding of individual components into a mechanistic understanding of the model's behavior as a whole. It would be very convenient if the natural subunits of neural networks - neurons and attention heads - are monosemantic - that is, each component corresponds to "a single concept". Unfortunately, by default, both neurons and attention heads seem to be polysemantic: many of them seemingly correspond to multiple unrelated concepts. For example, out of 307k neurons in GPT-2, GPT-4 was able to generate short explanations that captured over >50% variance for only 5203 neurons, and a quick glance at OpenAI microscope reveals many examples of neurons in vision models that fire on unrelated clusters such as "poetry" and "dice". One explanation for polysemanticity is the superposition hypothesis: polysemanticity occurs because models are (approximately) linearly representing more features[1] than their activation space has dimensions (i.e. place features in superposition). Since there are more features than neurons, it immediately follows that some neurons must correspond to more than one feature.[2] It's worth noting that most written resources on superposition clearly distinguish between the two terms. For example, in the seminal Toy Model of Superposition,[3] Elhage et al write: Why are we interested in toy models? We believe they are useful proxies for studying the superposition we suspect might exist in real neural networks. But how can we know if they're actually a useful toy model? Our best validation is whether their predictions are consistent with empirical observations regarding polysemanticity. ( Source) Similarly, Neel Nanda's mech interp glossary explicitly notes that the two concepts are distinct: Subtlety: Neuron superposition implies polysemanticity (since there are more features than neurons), but not the other way round. There could be an interpretable basis of features, just not the standard basis - this creates polysemanticity but not superposition. ( Source) However, I've noticed empirically that many researchers and grantmakers identify the two concepts, which often causes communication issues or even confused research proposals. Consequently, this post tries to more clearly point at the distinction and explain why it might matter. I start by discussing the two terms in more detail, give a few examples of why you might have po...

Since the late 1990s, the standard practice for GDM care has been to measure postprandial glucose values. For patients with pre-gestational diabetes, whether type I or type II, the ACOG recommends multi-level glucose checks (fasting, pre-meal , postprandial, and nighttime). But what about in the immediate postpartum interval? In patient's with pre-existing diabetes, should blood sugars be checked pre-meal (qAC) or postprandial while still in the hospital, and after discharge? The topic for this episode comes from one of our podcast family members who had this clinical dilemma? In this episode, we will review the data and recommendations from the American Diabetes Association, the ACOG, and CDC. So grab your sugar-free drink of choice, and listen in!

Gestational Diabetes (GDM) is vastly more prevalent in pregnancy compared to pre-existing diabetes. In 2009, the ACOG states that 7% of all pregnancies were complicated by a diabetes diagnosis, with 86% being GDM. The prevalence of GDM keeps rising in the US and globally. Metformin is increasingly prescribed in pregnancy, yet its long-term effect on the neurocognitive development of the offspring remains incompletely described. However, newly published data (March 6, 2024; AJOG) has changed that! In this episode, we will summarize and review a systematic review and meta-analysis of childhood neurodevelopmental outcomes after in utero exposure to metformin. Additionally, does some evidence suggest that metformin may be superior to insulin in pregnancy for perinatal outcomes? We will discuss all this and more, in this episode. This information will be helpful as we counsel and educate our patients on metformin use in pregnancy.

The “traditional“ Parkland management protocol for GDM included the immediate initiation of medical therapy for those with abnormal fasting blood sugar, in addition to another additional value, on the 3 hour GTT. These patients were automatically labeled as A2 GDM at time of diagnosis, rather than waiting the 1 to 2 weeks of nutritional/diet therapy. Does fasting hyperglycemia on the 100g GTT truly predict the need for subsequent medical therapy? In this episode, we will summarize new data on this subject from AJOG MFM published on February 17, 2024. Does immediate medical therapy after GDM diagnosis improve overall maternal/neonatal outcome? It's a complicated answer, and we will review it in this episode.

The ACOG has consistently recommended universal screening for gestational diabetes between 24 and 28 gestational weeks. Although controversial, the ACOG does endorse earlier screening for GDM in patients with additional risk factors. But what about patients who present for prenatal care after the 28th or 29th week? Should screening for GDM be done in the 3rd trimester? And if we do screen in the then, what is the reference range for “normal “or “abnormal”? Is it the same interpretation as when it is done between 24 and 28 weeks? Does 3rd trimester screening impact parental outcome? In this episode, we will examine the data and provide a recommendation of when testing for gestational diabetes in the 3rd trimester may have the most impact.

Gestational diabetes (GDM) is a risk factor for adverse perinatal outcomes. Currently, the ACOG recommends early screening for GDM for women “at risk”. However, other experts disagree with this approach. On October 6, 2022 we released a podcast episode called “Early GDM Screening: Evidence-based?”. In that episode we covered the controversy regarding early GDM screening, in other words- screening under 24 weeks. We have been following this story and debate for over 2 years now; we first released the episode investigating the utility of early screening back on May 7, 2021 with an episode called “early GDM screening: Does it matter?”. The controversy surrounds maternal and neonatal outcomes… does it improve with early screening? Well… we have more data now! YEP.. looks like we were vindicated in our prior messages covering this! In this episode, we will summarize key findings from a recent June 2023 publication in the NEJM titled, “Treatment of Gestational Diabetes Mellitus Diagnosed Early in Pregnancy”. The lead author is Simmons. So…should we be doing early screening for GDM? We'll highlight the data.