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 585 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
Altman: OpenAI made big mistakes in Hugging Face incident, an alignment failure
“Here are the mistakes that OpenAI made, and we did make some big ones, of course, but these systems have gotten incredibly capable. So I think it's an alignment failure. I think it's a security failure. I think it's like a very serious thing, even though it's,…”
Musk: OpenAI was founded to prevent concentration of advanced AI
“The reason that obviously you mean the rest of the team you know, created open AI was to help
[783] Elon Musk: with the democracy, help, help spread out AI technology
[787] Elon Musk: so it doesn't get concentrated in the hands of a few.”
Cherny uninstalled his IDE and lands 20 AI-written PRs daily
“For me personally, it's been a hundred percent for like since Opus 4.5. I just, I uninstalled my IDE. I don't edit a single line of code by hand. It's just a hundred percent quad code in Opus. And you know, I land, you know, like 20 pairs a day, every day.”
Scholl: Boom needs more software engineers post-AI because development costs dropped
“At Boom, we need far more software engineers in a post AI world. They would need it in a pre AI world. Why? Because the cost of software development has dropped. Anybody, including hardware engineers can now become a coder. And we need software engineers to ma…”
Altman: OpenAI must avoid imposing its worldview on AI users
“And that means that we have to, you know, be fairly quite reserved about not imposing our worldview about what we think people should or shouldn't do with AI, or even what we think the good ideas are, while still making sure that we can ensure enough of a safe…”
Cherny: Anthropic uses Claude routines to maintain its own apps via Slack
“We actually have Cloud maintaining itself now. And the way we do this is we have a Slack channel where we just had Cloud start a bunch of different routines to maintain its own code base. And we actually do this for the CLI, for the iOS app, for the Android ap…”
Hassabis: DeepMind makes edge models open because on-device models are vulnerable anyway
“For now, what we've decided is that our edge models, the things we want to use for Android, and glasses, and robotics it's best that they're open models. Because they're vulnerable anyway on the, once you put them out on the surfaces, so they might as well be …”
Massad: In 2023, Replit explicitly chose not to target traditional developers
“So at some point in 20, 23 we just made it an explicit goal of like, we're not going after developers.”
Vuong: PI's complex robot evaluations actually run on remote cloud models
“Almost all of the robot evaluation that we run at Pi today, including the really Complicated demo that we have shown making coffee, folding laundry, mobile robots navigating around. The model actually hosted in the cloud.”
Vuong: PI integrated with robot startups without ever inspecting their hardware
“And the, one other interesting thing about our collaboration with Weave and Ultra is one, I've never seen their robot in person. Two is I have very little idea about how their robot actually works. And that's a very intentional choice. I want to stay away from…”
Poetiq's meta-system generates reasoning systems for problems GPT-5 cannot reliably solve
“And so the core technology that we've developed at Poetic is recursive self-improvement. So we have a recursively self-improving system, which we call the Poetic meta system. The output of that system is systems that solve hard problems where a hard problem is…”
Anthropic's path to safe AGI centers on coding and tool use
“You know, as Anthropic, I think for Ant, the bet has been coding
For a long time.
And the bet has been the path to save, to safe AGI is through coding.
And this is, this has kind of always been the idea.
And the way you get there is you teach the model how to …”
Garry Tan grants Claude Code direct access to his production database
“I mean, I'm not sure if I'm supposed to do this, but I've actually also had it access my production database.”
Heller: Casetext built CoCounsel with plain Python, dismissing frameworks like LangChain
“And to be honest, a lot of the stuff that we built while building co-council was exactly like this. Every time you do this task, you're basically gonna take the same six or seven steps. And you don't need to have, frankly, like fucking Langchain or whatever. J…”
Srinivas: Perplexity's primary strategic bet for the future is an AI browser
“The browser. That's the big bet we are making as far as the future of the company goes.”
Ng's teams rebuilt a codebase three times in one month using AI
“So I'm on teams where, you know, we've completely rebuilt the code base three times in the last month, right? Because it's not that hard anymore to just completely rebuild the code base, pick a new data schemer, it's fine, because the cost of doing that has pl…”
Ng: One project team proposed staffing twice as many PMs as engineers
“So literally yesterday, one of my teams came to me, and for the first time, when we're planning a head comfort project, this team proposed to me not to have one PM to four engineers, but to have one PM to 0.5 engineers. So the team actually proposed to me. I s…”
Altman: OpenAI has an open-source model releasing soon
“We have an open source model coming out soon.”
Musk: xAI deploying 140K GB200s and 150K H200s in Memphis
“So now we've got 150,000 H 150 K H 200 and 30 K GB 200 in the Memphis training center. And we're about to bring a 110,000 GB 200 online at a second data center also in the Memphis area.”
Truell: Cursor's ultimate goal is to replace traditional coding entirely
“Yeah, the goal with the company is to replace coding with something that's much better.”
Truell: Cursor bans AI tools during its initial software engineering technical screens
“So for interviewing, we actually still interview people without allowing them to use AI other than autocomplete for our first technical screens.”
Tan: Y Combinator uses LLMs internally to evaluate prospective venture investors
“Well, one of the things we've been using LLMs for internally is actually helping founders figure out who they should take money from.”
Altman: OpenAI now knows the path to building AGI
“This is the first time ever where I felt like we actually know what to do. Like, I think from here to Building an AGI will still take a huge amount of work. There are some known unknowns, but I think we basically know what to go do, and it'll take a while.”
Seibel: Ignore investor rejection reasons until hearing them across 15 to 25 pitches
“When founders are fundraising, it's like the first thing, the first time they tell me there's a rejection and a why, I ignore it completely. If they've gotten like 15, 20, 25 rejections, I then ask, are the whys consistent? And only if then the whys are consis…”
Benjamin's team plans to scale using networked five-qubit modular units
“We actually are planning to build a machine, which will be made out of modules, which each one is a small quantum computer, and by the way, so small that on its own, it's not good for anything. Even as few as five qubits, right? And then we have modules, and w…”
Altman: Best investments either lacked other investors or felt incredibly expensive
“The very best investments I've made have either been a ridiculous deal because no one else wanted to invest, or a deal that felt incredibly expensive. The more I was willing to sort of overpay in my mind to invest in a company, often the better it did.”
Altman: YC strongly prefers multi-founder teams with a technical founder
“Where I've seen that go wrong the most often is then the ability to attract, evaluate, and retain technical talent. And so, we have a strong, strong preference. For founding teams that have at least one technical founder. We also have a strong preference for t…”
Allen Used Fake Accounts to Simulate Activity on Indie Hackers
“The first week created a bunch of fake accounts, which I had heard other people doing. I was like, all right, this seems cheesy, but you know, I'll try it. And I would have conversations with myself.”
Zapier pays $10,000 to new Bay Area hires to relocate elsewhere
“If you're in the Bay Area and you take a job with Zapier, we'll pay 10,000 dollars to help you relocate to anywhere, or so we call it delocate.”
Scholl: Jeff Bezos passed on Boom's 2015 seed round
“One of my very many failed investor pitches was in 2015 when Jeff Bezos passed on our seed round at Boom. And then he wrote the 2015 Amazon shareholder letter about how there's certain things that only big companies can do. And he cited airliners as one of the…”
Huang: NVIDIA physical AI and automotive business is near $10 billion
“Our robotics business, autonomous vehicle business, basically physical AI business is probably almost, it's like ten billion dollars, so it's really, really big already.”
Garry Tan discloses spending $1M annually on AI tokens
“I'm spending a million dollars a year on tokens.”
Brex Uses AI Agents to Automatically Fix Other Agents' Bugs
“Whenever someone has a conversation with the agent that is the flags an issue or a bug or something that feels like the conversation didn't go as smoothly, that creates a bug. That bug triggers an agent that's going to go and modify the code base and the promp…”
Koomen: YC runs a nightly AI agent to review employee interactions
“We have this general agent that every night will go, And read through all of the agent conversations that employees have had, and look for things that could have done better, and pieces of context that if it had, up front, it would have done more efficiently.”
Tan: Building Garry's List with Claude Code took $200 and five days
“The first time it took about, you know, four million dollars and, you know, six or seven people and about a year and a half. And then the second time it, you know, took about, I don't know, a hundred grand and two people, me and my co-founder Brett Gibson, who…”
Garry Tan no longer uses Visual Studio, relying entirely on AI agents
“For like, I don't use Visual Studio at all. Like there's no reason to, like when I can talk to my agent and my agent can do this, right?”
Vuong: Claude-based on-call agent boosted compute utilization by 50%
“We have a Claude skill that essentially serving the role of a pre-training on call today. So, you know, we have these pre-training runs that are really large it's very, I think, a difficult exercise to keep them alive, to, you know, for them to continue to chu…”
Chollet: NDEA Has Only a 10% to 15% Chance of Success
“Like we have maybe a 10 or 15% chance of success. But that is enough that it's worth trying, right?”
Chollet: ARC-3 private test set differs substantially from public set
“We've deliberately tried to create a private set of environments that is significantly different from the public set. Like you can look at the public set. It's not actually giving you that much information about what's in the private set. In the private set, y…”
Emergent agents learn across user sessions using long-term trajectory memory
“We were able to figure out, okay, all the trajectories that we are generating, we can kind of aggregate over time and like sort of build in a long-term memory for the agent, which is very unique in the sense that your agent learns not just from your own sessio…”
Hodak: Science has universal stem cells hidden from the immune system
“So we there's, it's one cell line and the, probably the single deepest area of IP on this is that we've hidden them from the immune system. So we're one of a really small number of companies that have, I think like pretty convincing, what we call hypoimmunogen…”
Hodak: Personal probability of AI doom is well below 50 percent
“Like I don't, my PDOOM is pretty, it's not zero, but it's not 50%. It's well below that.”
Claude Code stayed a CLI because graphical UIs obsolete too quickly
“And really, I think that's why we stayed in the CLI is because we felt there is no UI we could build that would still be relevant in six months because the model was improving so quickly.”
Anthropic uses Claude Agent SDK to automate reviews and deployments
“We use the Quad Agent SDK to automate pretty much every part of development. It automates code review, security review it labels all of our issues. It shepherds things to production. It does pretty much everything for us.”
No code in the Claude Code codebase is over six months old
“All of CloudCode has just been written and rewritten and rewritten and rewritten over and over and over. We unship tools every couple weeks. We add new tools every couple weeks. There's no part of CloudCode that was around six months ago. It's just constantly …”
PostHog is building a desktop app that autonomously fixes bugs via PRs
“At the moment we're working on like a desktop app, for example, that ships pull requests based on your customer data. So we'll look across all your session recordings or your analytics or your error tracking, your LM traces, go, Hey, these are the issues we ca…”
Mignano: Lightspeed got lucky timing Suno investment and may have overpaid
“You know, with Suno, I feel like we kind of got lucky that we met the team at a certain time and, ah, were able to invest like kind of just as it was inflecting, and Yeah, sometimes you just get lucky and maybe overpay just to be able to do it.”
Anthropic built custom distributed training to scale beyond Facebook's infrastructure
“We don't want to outsource this to some package because A, we're about to go to a bigger scale, like PyTorch, for instance, they had a package for doing this. But we were going to go to a bigger scale than Facebook had been to. And you don't want to have a dep…”