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 993 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
CB Insights sees a bigger opportunity competing with McKinsey than Bloomberg
“I, you know, I really think of us more, to some degree, going after, like, McKinsey, right? So if you think about McKinsey services as a pyramid, there's sort of that bottom layer, which is sort of, you know, what are my competitors doing, and where are market…”
Mortensen: x.ai has no backup plan to rely on human labor
“What we kind of set out to do is to kind of create a setting for where there is no plan B. There is no kind of backup plan, and I know SoftBank kind of, you know, shit their pants whenever I say that, but there is no plan B of, ah, perhaps we can just have hum…”
Pinterest data scientist John Rauser cannot derive the t-distribution from scratch
“There might be five people in this room, the people who raised their hands, who could sit down and just derive this from first principles. I am certainly not among those people, and I am a working data scientist.”
Atlas Venture will not fund startups rebuilding existing software platforms
“So for my money, I don't want you to spend my money rebuilding, even if it's a little bit better, something that's been built. I want you to spend my money taking something that's been built that's good and disintermediating those guys.”
Wilson: USV will not invest in porn, spying, or gambling
“I won't invest in porn. I won't invest in spying. I won't invest in gambling. Those would be three things I wouldn't do.”
Wilson: User value and team quality matter more than early business models
“I don't care about the business model because we can tweak the business model. We can get the business model right. What I care about are do you have users and are they getting value out of your product and how much value they're getting out of your product an…”
Lonsdale avoids poaching Palantir talent until the company's growth slows
“I, I'm not poaching from Palantir yet. I always told Palantir, if they stop growing really quickly, I'm going to take their best guys, but I can't do that yet.”
The CFPB approached ZestFinance offering regulatory protection for algorithmic underwriting
“We've, ah, we've actually been approached by the Consumer Financial Protection Bureau, ah, about, in essence, giving us kind of a waiver to say that, you know, we're gonna protect you from regulators who dispute this or don't like it.”
Matt Ocko: Data Collective got 2x-3x cash returns on failed investments
“In three years, in very, very high risk investing, our worst case outcome was 95% return on capital. And in the failed companies, we got two X, three X cash on cash returns.”
Matt Ocko: Data Collective backed ZenPayroll to gut legacy incumbent ADP
“We're investors in a company called Zen Payroll, ah, and we're investors there because they demonstrated in, on a very small amount Really, of their own money, that they could gut ADP. That customers deserted ADP in droves to have essentially single click, ove…”
Socher: Recursive will build AI-for-AI before tackling physical sciences
“So at Recursive, we are fairly sure that we have to start with AI for AI and then make it really, really good at doing research on creating better AI so that it has the equivalent of 50,000 PhDs in terms of knowledge and its own capabilities, and only then go …”
Wolf: Hugging Face Used Quantized GLM 5.2 Against OpenAI Intrusion
“The model we use to counter open AI intrusion was GNM 5.2 that was quantized by Nvidia in, in, in four bits.”
Feldman: Cerebras burned $8M monthly for 18 months before building successful chips
“And we had a, an 18 month period where we were spending eight million a month and we couldn't build them.”
Katti: OpenAI considers designing and building its own data centers
“As we go forward obviously there'll be building on all of these relationships but also looking at more options where we design the compute, the data center itself, also potentially even build the data center ourselves.”
Prince: Cloudflare viewed early VCs strictly as money sources without wanting advice
“What our style tended to be was we thought of venture capitalists primarily as, you know, sources of money and, you know, mostly sort of, we wanted when we needed money, we wanted to be able to access capital, no thought leadership, but we, yeah, we really jus…”
Prince: Cloudflare's initial internal AI agent was secretly staffed by 20 humans
“So for us, it was auto at cloudflow.com. And we told everyone, it's a magic AI agent. We've set it up. You can ask it anything. Now, in reality, behind the scenes, it was a team of about 20 people that staffed this 24 seven.”
OpenAI plans to increasingly rely on reinforcement learning to scale intelligence
“When you have a lot of compute, you want to turn that compute into intelligence in a way that's useful, and RL is one way of doing it, and we just started doing it then, and we're going to do a lot more of it now.”
Dubois: OpenAI expanded RL training from math competitions to real-world coding
“We were able to take many of the tools that we built for these, like, verifiable reward cases, and we were able to use them more generally in on, for reinforcement on, like, real use cases, and I think that's, like, really why we're feeling that right now in, …”
Dubois: OpenAI internal sentiment goes through waves of hype and doubt
“It's kind of funny because in general with every model that is looking really good early on we have a model, we all get really excited about it. And then there's like tons of doubts. That start coming up because it's like, oh, like everyone is so high, is like…”
OpenAI's Yann Dubois rarely uses GPT-5.5 Pro due to high latency
“I personally don't use Pro that much because I really don't like waiting. I'm pretty impatient, so I don't like waiting for that long, and I know that the probability of being correct definitely improves, but it doesn't improve, like, enough for me to use it.”
Burazin: Client requested 5 million concurrent AI agent sandboxes
“But we have a request today for five million concurrent look. So five million at one point in time, right?”
Burazin: Sandboxes running over 24 hours represent 2.5% of volume, 20% of revenue
“Two and a half percent of our sandboxes run longer than 24 hours, but it's a large percentage, like twenty-ish percent of revenue.”
OpenAI's Safety Committee has authority to delay model releases
“In the case essentially where we have more questions, we can delay model release if we feel that we need to understand that, that better.”
Kolter: Calculating a P(doom) for AI is a flawed concept
“I have never expressed a P. Doom and things like this. I just think it's a very weird concept as if the world is some stochastic set of dice that you can roll multiple times and that we don't have direct influence over this.”
Anthropic launched Project Glasswing to help critical infrastructure maintainers fix vulnerabilities
“Yeah, so Project Glasswing is a project that is attempting to give the people and the companies that provide much of our software infrastructure sort of the very foundation. The Linux Foundation is an example that is pretty close to my heart as a member of the…”
Rieseberg: Anthropic Is Prioritizing Local Execution for Claude Cowork
“So in the short term, I want to make it very possible for Claude to meet you where you're working. If you're working on your local computer, that's where Claude should be.”
Safe execution environments are currently the bottleneck for autonomous AI development
“Like in a safe setup, you know, where they can like, because right now we definitely, we don't, we're not confident about, you know, them doing the right things all the time and measuring like how much they can push and how long they can push a task is very di…”
Dehghani initially thought Google's Transformer architecture was random and would die
“And I was like, I don't know if I want to go with this team. It's just like, they're doing something random. Like who, like everybody's doing LST. I'm like, why should I go and work with like a group of people who are working on this like random architecture, …”
LangChain adds tools for AI agents to autonomously trigger context compaction
“One interesting thing there, actually, that we haven't yet released as of this recording, but will probably be released by the time it comes out, is we actually give the agent a tool to trigger its own compaction.”
AxiomProver self-improves by adding generated mathematical proofs to its skill library
“Action Prover learns to prove things, and it kind of self-improved in a way where all the things that it proved got fed back into it, into a kind of a skill library.”
LeCroix: Mistral 4 development is compute-constrained; deploying Grace Blackwell capacity
“Definitely compute and the current deployment that we have will help as it's going to be giving us a lot more Grace Blackwell capacity than we had in the past.”
Raschka does not use LLMs to write his books or blog posts
“For blog writing on book writing, not so much because honestly, I, for fun, I tried it out. It's just, it generates okay text, but it's, I don't know, it does not I can ask it to generate text like me, but it's almost like, then I don't like it and I end up ed…”
DeepMind's Bourgeau: AI progress is ahead of where I expected
“I think, if I'm being honest with myself, I think we're ahead of where I thought we could go.”
Ai2 fine-tuned OLMo 3 using Chinese teacher models DeepSeek-R1 and Qwen
“So in our case, we took a mix of existing data sets like Open Thoughts three and modified it, which is from Bespoke AI labs, a startup. And then we also generated a whole bunch of new data. So we ended up using a mix of teachers from like Deep Seek R one, oh f…”
Lambert: Ai2 generated billions of DeepSeek completions over a weekend
“We had a bunch of cloud credits and I, they were running out and we're behind and I just generated like as many completions as possible. So it was like a few billion completions from deep seek over the weekend.”
Kant: Poolside's Project Horizon is an expandable two-gigawatt data center campus
“So Project Horizon, in a nutshell, is today announced a two gigawatt campus. We can actually go far beyond two gigawatts on the site that we're in.”
Kant: Poolside starting construction on 250MW natural gas data center in 2025
“And so what you're seeing from us is that before the end of this year, we're starting construction on a 250 megawatt data center. It's natural gas powered with grid connection as a backup, and it will house an incredible amount of compute, but it will be the f…”
Kant: Poolside is building a 250MW data center with under 450 workers
“When we're building 250 megawatts, We're doing it with a less than 450 person on site workforce with very skilled tradesmen and construction workers that we're bringing over there and we're housing over there versus a world where you might need traditionally 2…”
Tworek: OpenAI's high and low reasoning modes use the exact same model
“Where you can have like a high rezoning model and the low rezoning models. And this is like in the end, the same model. You just, we just tweak the parameter, which says we want you to think longer or shorter.”
Tworek: OpenAI focuses research on only three or four large-scale projects
“We all work on a very few projects total. There are not that many projects. OpenAI is not trying to do everything. We are not trying to like have portfolio. We are trying to have like multiple different bets. Always the idea is we do a few core things really, …”
OpenAI VP Jerry Tworek pays $200 per month for ChatGPT Pro
“I think I am pretty heavy user of ChatGPT right now, happily paying like 200 dollars a month for it”
Tworek: OpenAI's RL algorithm is not GRPO but shares similar components
“Like what we, what OpenAI is doing is not exactly GRPO. It is slightly different in many different ways, but like some parts are definitely similar.”
Tworek: OpenAI does not currently train models online via live interactions
“This is not what I am aware, at least, like, not, not, not what OpenAI is doing at the moment”
Douglas: Anthropic focuses on alignment and near-term economic coding utility
“Anthropic has been laser focused on on two things. One is, like, long-term AI alignment, and two is near-term economic impact. So Anthropik has been laser-focused on coding and computer use, and things that we think will make a direct impact to the economy, li…”
Douglas: Anthropic intentionally deprioritized math reasoning unlike OpenAI and DeepMind
“You know, one thing that Anthropik, like, You noticeably hasn't focused on compared to DeepMind and to OpenAI is is mathematical reasoning, right? DeepMind and OpenAI have been pursuing mathematical reasoning because of the implications for science and for sci…”