The Ledger
Every statement that passed quotation and attribution checks. Mix any filter with any other: certainty 1/5, debate potential 5/5, or both at once.
why aren't all 1,046 resolved? a statement only gets an assessment when the public record can support or contradict it. opinions and what-ifs never can, and 0 checkable ones are still open, waiting for their date. predictions held up or didn't; assertions are supported or contradicted. on every card: ▮▮▮▮▮ certainty · ▮▮▮▮▮ debate potential. speakers are clickable
FourCastNet matches supercomputer weather accuracy 10,000 times faster on consumer GPUs
“To our surprise, we found that it's not only, you know, accurate, it's almost as close to the what the traditional weather models can do accurately, but also tens of thousands of times faster. So what would take a big supercomputer to run can now be run. And w…”
FourCastNet predicted Hurricane Lee's landfall days earlier than standard weather models
“For instance, our forecast net was able to correctly predict that the hurricane making the landfall several days earlier compared to the standard weather forecasting models.”
Spherical AI weather models achieve longer autoregressive rollouts than flat models
“These models that we have are able to do the longest rollouts compared to any of the other weather models that completely ignore spherical assumption and a range of other things.”
FourCastNet models trained on six-hour steps produce stable months-long forecasts
“To predict for the next six hours. And a little bit of multi-step fine tuning... Now we are showing for several months that it's able to do that.”
Park: Training on randomized controlled trials improves AI human-behavior prediction
“That by collecting a lot of these randomized control trials that are really well designed, we can make significant improvement in models capability to predict human behaviors.”
Krentsel: OpenClaw, Pi, and Claude Code hardcode static agent policies
“These are all policy decisions that are static, that are defined for OpenClaw, or for Pi, or for Clawed code, if you look at their Source code. And so that is the kind of, that is the policy of what an agent is, the tools it can use, the skills it has, how it …”
McPartlon: Chai-2 generated binders for 25 targets with 20% hit rate
“So we designed antibodies to 50 targets for that paper, got binders to about half of them with being on average around a 20% hit rate for binding.”
McPartlon: Chai-2 achieved 0.33 angstrom error in cryo-EM testing
“And in this case, it was a 0.33 angstrom error, which is one third the width of an atom.”
McPartlon: AlphaFold-Multimer gets antibody-antigen prediction right only 11% of the time
“Not really like outfold to got like, I think, 11%, the multiple version of this got like 11% of antibody antigen prediction cases. Correct. That means 90% of the time it's wrong.”
Patil: Chai-2 demonstrated precise antibody-based GPCR agonist activity
“For example, in CHI-II, we showed like GPCR agonist activity, right? Where you can really hit the switch on a, you know, on a cell doorbell protein, so to speak, right? In a very precise way. Very, very, very hard to do that with antibodies if you can't be tha…”
Kant: Poolside's 8B-active Laguna S solved Erdős 397 on DGX Spark
“A 118,000,000,008 B active model, which is not that large. It fits on a DGX spark and still runs at, you know, 3040 tokens a second on a spark is able to solve. Erdos three 97 independently. It's able to do complex programming tasks.”
Chu: X-Cell Predicts Perturbation Effects In Unseen Activated T Cells
“Critically, Excel has not seen active cell T cells. And it's able to make accurate prediction, not only on the known biology, the TCR complex, predicting their effect accurately, that these are going to inactive the T cells, which It's exactly what we would ex…”
Chu: Xaira is first to combine seven genome-wide Perturb-seq campaigns
“It is the first time that someone can put together not just one perturbseek, but seven genome-wide perturbseek campaigns together.”
Bubna: Open-source DFlash matches proprietary speculative decoding performance
“Recently we shared our work on dflash, which is a block-based speculator, and we've open sourced all of it, so you can get, by using open source dflash, you can get the same performance as you would with one of the proprietary providers.”
Cohen: OpenClaw logs all messages in plain text
“I started to see the size of the code base and the number of dependencies and some other things like logging all messages in plain text that just made me a bit apprehensive to use it for like production use cases to build a business on it.”
Xin: Every major analytics database engine in traction is a decade old
“Actually, every single database engine out there, especially on the analytics side, are kind of a decade old. Pretty much everything that had reasonable traction are about a decade old.”
DeepMind, Microsoft, and Meta are building or using physical science labs
“You see people like Google DeepMind, Microsoft, other places like Meta, either building their own lab or running experiments at someone else's lab to get that data back.”
Andon Labs AI agent Luna published job listings and hired human employees
“So it has two, two people that it hired. It did job listings.”
Hong: DeepSeek dissolved its formal reasoning team over strategic shift
“And we have since, for example, Deep Seek All right. Like originally having a formal team and then later dissolve that team because of strategic direction change.”
Hong: Harmonic's Aristotle Verified an Erdős Problem Proof Found by GPT
“In fact, like, you know, GPT found a proof to an unsolved Erdos problem, and our competitor Harmonic, you know, Aristotle you know, verified it.”
Ethan He: Storing and moving video datasets costs millions per month
“So, so it's like just storing, storing the network, those costs, it's just I guess it would be a few millions per month to just storing everything, not to mention the GPU costs.”
Sparse autoencoders found a single learned feature for nucleophilic elbows in ESMC
“You know, what we found basically is that the model has a kind of a single feature for this nucleophilic elbow and is activating across these like very evolutionarily diverse families, you know, really completely different structural topologies, proteins that …”
Incorporating metagenomic training data eliminated diminishing returns in ESMC protein foundation models
“And then, you know, what we saw basically is, is, is there are no longer diminishing returns to scale. So that's really saying that ESM two was kind of data limited rather than compute limited for ESMC.”
Rives: ESMC Has Been Used to Successfully Design scFv Antibodies
“We've been able to use this to actually now go and design many protein binders, but I think sort of most excitingly, we've been able to use this to actually design antibodies, SCFVs, and we're seeing really, I think, exciting success rates and a small number o…”
Burazin: Daytona built a Windows sandbox that spins up in one second
“And the only option right now is an EC two with Windows or on, on, on Azure.
Both of them take anywhere from three to five minutes to spin up.
We've created an actual sandbox.
So it's a second instead of milliseconds, but you have like point in time snapshots,…”
Azhnyuk: Standard 10-Inch Ukrainian FPV Drones Strike at 30-40 Kilometers
“Many hits are happening between 30 and 40 kilometers, and that's what expected from a regular 10 inch, ah, FPV drone.”
Azhnyuk: Ukraine made 4M FPV drones last year with 7M targeted this year
“Last year alone, Ukraine manufactured about four million of these, and then Russia's maybe, like, 20% less than that. And for this year, the publicly voiced target was seven million on the Ukrainian side.”
Azhnyuk: Inexpensive local hardware now runs near-SOTA open-source AI models
“You can have basically an inexpensive computer running, you know, what was a state of the art model year and a half ago, running it locally on a device with an open source model, which also means that the Chinese can have it, the Russians can have it, the Nort…”
Smith: Success rate of large powers invading smaller states has dropped
“They fail a lot more now than they used to. The success rate of taking places over has gone way down.”
Applied Intuition operates driverless L4 autonomous trucks in Japan
“We run like, as an example, we run driverless trucks in Japan right now, like as we speak, we can't have errors that are L four trucks.”
Swix: YouTube's recommendation system is LLM-based using tokenized video codebooks
“The YouTube Rexxus is LLM-based, and they- Is it really? They obviously, yeah. That's cool. It, they re, they tokenize every video, and put it in a code book, and then they train a LLM on it, and then feed in your context, just like a regular LLM, and ask it t…”
D'Amico: Decoupling appliances from the grid via batteries is unprecedented
“If you slam a battery into it, you're now not, you, you've decoupled the energy input from the wall with the device's power outputs. You can decouple the user experience from the grid. And that level of, like, approach has not been done in kind of the major ho…”
D'Amico: Impulse cooktop boils a liter of water in 40s with 10,000W
“So we're putting 10,000 watts of power into a, like, medium-sized pan. Or pot. And, ah, we can boil about a liter of cold water in about 40 seconds.”
D'Amico: Impulse stoves add a 3kWh battery for 15,000W boost power
“So basically we put a three kilowatt hour LFP battery, lithium iron phosphate inside home appliances, starting with stoves that gives us an additional 15,000 Watts of additional power. So that's like double what a typical induction cooktop has in terms of aggr…”
Materials Project and Open Catalyst datasets rely on low-fidelity DFT calculations
“Materials project open catalyst project, these do provide good leaderboards, but some of the limitations that are the data comes from not very high fidelity density functional theory. So I'd say that's a second challenge is that we're all, all the smartest ML …”
Colvin: Deno cannot control memory usage and prevents OOM protection
“Even if you don't allow that Dino does not have any way of controlling memory. So even if someone can't run arbitrary code, they can oom your machine as often as they like.”
Colvin: Monty executes Python in single-digit microseconds
“I mean, in a hot loop, we can run, we can go from code to execution result in under a microsecond in like 800 nanoseconds. In reality, it's like single digit microseconds to run code or single digit microseconds to run the next step of a REPL or single digit m…”
Google developers internally use a coding tool called Jet Ski
“They use an internal thing called jet ski. And I mean, the simple reason is they have internal versions of everything else. And so if they actually train the, they release the internal version to the external world, it would just make no sense. Like it would j…”
Eskildsen: Turbopuffer ANN v3 searches 100B vectors with 40ms p50 latency
“ANN v. Three can search a hundred billion vectors with a p-fifty of around 40 milliseconds and a p-ninety-nine of 200 milliseconds. Maybe other people have done this. I'm sure Google and others have done this, but we haven't seen anyone at least not in like a …”
Eskildsen: Cursor moved 20 terabytes from Postgres to Turbopuffer
“Like at some point, cursor moved like 20 terabytes of Postgres data into Turbo Puffer, because it's like, it's there, it works, and these particular query plans we know work well, and so they just moved it all to defer sharding.”
Shah: Supermemory outperformed Claude Code and OpenClaw benchmarks by almost 50%
“So the Claude code one performed the worst, and OpenClaw slightly more than that, and SuperMemory is the highest, and you can see that, you know, it's like a pretty significant difference, like almost 50%.”
Shah: No other memory provider offers hybrid memory and raw RAG fallback
“Essentially, you will return the memories first, and then if there's any raw chunks that match up, we also return those to make sure the agent knows just enough information to answer the question. And no other provider does this right now”
Nelle: No one had enabled AI coding agents to run code before Cursor
“Like obviously you need to run the code. And so that I think also is probably not that contrarian of a take, but no one has done that yet.”
Huber: Frontier AI models are not actually good at agentic search
“We've like sort of stress tested like frontier models and their ability to search. And they are not actually that good at searching.”
Levie: Anthropic has forward-deployed engineers embedded at Goldman Sachs
“OpenAI probably is hiring FDEs to go into the enterprise and then Anthropic is embedded at Goldman Sachs.”