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 197 resolved? a statement only gets an assessment when the public record can support or contradict it. opinions and what-ifs never can, and 7 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
Swyx predicts coding SaaS will outlive all other SaaS
“Coding SAS will be the last SAS in the world because all the other SAS will just be buildable via coding SAS.”
Google's internal compute credit market caused it to miss GPT
“This is a thing that has been tried internally within Google, and it led to Google missing GPT.”
Swix: Pull requests and code reviews are dying, replaced by prompt requests
“The pull request is dying. Yeah. Right? It's going to be the prompt request. And then beyond that, code review is also kind of dying”
Swix: Standalone Vector Databases Are Over as a Category
“Everyone has vector indexes now, but like, I think it's fair to say vector databases as a standalone category are over.”
Swix: OpenAI will issue a cryptocurrency token to fund compute
“There is still one more shoe to drop, which is the non sovereign wealth funding that open AI needs to get, which they've promised to drop by the end of this year.
And my money is on, they have to do a coin.
Like it's, I'm not a crypto guy at all, but like, y…”
Swyx: AI models hit a 2T parameter wall, won't reach 10T
“As far as everyone is concerned, Claude, you know, Opus 3.5 is not coming out. GPT 4.5 is not coming out. And Gemini two, like we don't have pro whatever we've hit that wall, whatever that wall is. Maybe I'll call it like the two trillion parameter wall. Like …”
Swyx found Codex preferable to human marketing consultants for ads
“Codex has been my CMO for the last like two months. Basically running our ads and giving me advice on AEO and SEO and what have you. And I find it much more preferable to talking to a real human consultant, which I've also done. And it's roughly the same resul…”
Custom $50B ASICs become economically rational for $500B AI training runs
“If you have a trillion dollar or five hundred billion dollar training run, then take fifty billion of that and make an ASIC. Like, it's fine. Like you will get more than 10% efficiency from the ASIC. And like, that makes sense.”
Swyx: AI p(doom) over the next ten years is near zero
“I mean, if you do it in 10 years is near zero.”
Swyx: LLMs will plateau and potentially trigger a 30-year AI winter
“Yeah, probably LLMs are going to run out at some point and they're not AGI and okay, we have maybe another 30 years of AI winter or something and then like the next paradigm really is actually the thing.”
Swyx: Specialized agent labs like Cursor, Cognition, and Harvey will endure
“There will always be capability overhangs. They may not stay still, and so you gotta be nimble. But the Sierras of the world, the Cognitions of the world, the Cursors of the world, the Decagons and Harvey's, these are all agent labs for their field. They can b…”
Swyx: Agent labs will capture higher margins than model labs
“I think what I've been calling agent labs, which are people who build on top of all the other models. We'll probably have a better time with the margins because they price against the end user hours spent or like human labor. Whereas models get commodity price…”
OpenAI Plans to Scale Compute Power Capacity to 125 Gigawatts
“For OpenAI to go from like two gigawatts of compute this year to 30 with everything they've already announced, and then there's a plan for the next 125. Like, the United States uses 300.”
Swix: Frontier model sizes have likely peaked around 2 trillion parameters
“I wonder if we've hit the peak big model craze because now I do expect, you know, 10 trillion model releases, you know, 100 trillion model releases. Probably not. I think we might have peaked at two.”
Swix: Every frontier lab now distills dense models into MoEs
“I think like, I think this is the pattern for every frontier lab now.”
Swyx: AI inference providers are a proxy for the open-versus-closed model gap
“The other thing I always think about from a finance point of view that, Maybe people don't even think, don't think about that closely, but obviously push back if you disagree, is that it's basically a proxy for the gap between open models and closed models. Be…”
Swix: OpenAI Deep Research was built by three people as an o3 wrapper
“As far as I know, it's three people did it. It was Isa and like the two other collaborators that she had. I don't know if they did that much on top of all three, like every indication I've had from over the eye is that deep research is more or less a thin wrap…”
Swix: Cost-Unconscious Builders Win in AI by Building Ahead of Cost Curves
“So you should actually build your products ahead of where costs are so that by the time they are popular, actually your frontier. So it's actually, I think the efficiency thing trips up a lot of people in terms of Being cost and cost conscious and wasting a lo…”
Swix: Claude 3 degraded in capability a month after launch
“I used the same project to do this, to try to repeat the demo that I made for myself a month afterwards, and it wasn't anywhere as smart. So cloud three got dumber, but it looks like I like made up the demo or something, but no, like literally I just reran the…”
Swix: Chat completions are dying and early AI frameworks will age poorly
“Like, I don't think people realize, but, like, I'll just say it out here, like, chat completions is dying. So, like, any framework that was built in that era with, like, no conception of real-time, no conception of omnimodal or multimodal native things, they w…”
Swix: Descript still sucks despite major funding and OpenAI backing
“Descript is so much funding. They had OpenAI invested in them, and they still suck.”
Swix: Most podcasts are bad because creators lack real-world experience
“The reason that most podcasts or YouTube videos are shit is they're made by people who don't have life experience, who are not that important in the world. They're not doing important jobs. And so what you want to actually enable is CEOs to each of them make t…”
Swix: Deep research agents are the first agent category with true PMF
“What are the hard problems in this brand of agent that is like probably the first real product market fit agent. I will say more so than the computer use ones. This is the one where like, yeah, people are like, yeah, easily pays for 200 dollars worth a month w…”
Swix: AI engineering exists because labs crowdsource emergent capability discovery
“The reason that AI engineering can exist outside of the model labs is because the model labs release Models with capabilities that they don't even fully know because you never train specifically for it. It's emergent. And you can rely on basically crowdsourcin…”
DeepSeek-R1 researchers found MCTS and Process Reward Models were not useful
“R-one specifically said, yes, we tried MCTS. Yes, we tried PRMs. And none of that is useful.”
Swyx predicts OpenAI will launch a $2,000 per month ChatGPT tier
“I think that 2000 dollars ChatGPT will come.”
Swyx: VC appetite for GPU-rich early-stage startups is completely gone
“The appetite for GPU rich startups, like the, you know, the funding plan is we will raise sixty million and we'll give 50 of that to Nvidia. That is gone, right? Like no one's pitching that. This was literally the plan, the exact plan of like, I can name like …”
Swyx: The 'GPU smiling curve' rewards infrastructure and user-facing apps over middle-tier models
“I kind of call this the GPU smiling curve where the edges do well, cause you're either close to the machines and you're like number one on the machines, or you're like close to the customers and you're number one on the customer side. And the people who are in…”
Swyx: DeepMind has roughly a four-year advantage over OpenAI in world modeling
“So like they have maybe four years advantage on world modeling that OpenAI does not have. Cause OpenAI basically only started Diffusion Transformers last year when they hired Build Peebles. So DeepMind has a bit of advantage here.”
Shawn Wang says Devin's breakthrough was Agent-Computer Interfaces, not advanced planning
“The planner is like actually pretty simple, but ACI. That they book through on.”
Swyx says a16z-backed clipping tool Overclip wasn't very good
“We tried Overclip. I think it's the A-sixteen Z backed company that does clipping. It wasn't very good.”
Swyx claims Airtable founder Howie Liu had already sold the company
“I was also mentioned, I was also thinking about Howie Lu. From Airtable. Effectively just did the same thing with Hyperagent, except that he didn't run it in parallel that much. He basically had already sold the company and was just kind of doubleheading for a…”
Swyx: ChatGPT Work is OpenAI's most successful launch, surpassing GPT-5
“I have pretty much said this is the most successful launch in a long time. I think even more successful personally than five point O”
Swyx: Etched is not disrupting NVIDIA because inference requires dedicated ASICs
“I don't think they want to disrupt NVIDIA. I think the better framing is that all of inference is just so goddamn big that, of course, you're gonna have ASICs for inference.”
Swyx: Betting on transformer ASICs is safe because architectures persist
“The ChatGPT, like, GPT-III.V-ish architecture has mostly stayed the same this entire time. It's a pretty good bet, man. Like, even if there is a new, Model sometime in the future. The current workloads on existing models, like, four-oh is still being used, rig…”
Swyx: AI is too volatile for utility-style government regulation
“I think this thing is too volatile to treat as a utility, right? Like imagine if like Edison was like, you know, working on his like electrical stuff and light and Lighting and all that. And you immediately try to regulate it. Like, no, you'll probably wait 50…”
Swyx: AI p(doom) over the next 50 years is five percent
“50 is like, that's like the end of our lifetimes. And like, I think, I don't know, I'm gonna just throw out 10. No, ten's too high. Five percent.”
Swyx: AI engineers have permanent job security closing model capability overhangs
“AI engineer exists in the white surface area between the peak capability and deploying it everywhere else, right? So the more model research peaks and spikes capabilities in one domain, but it's not evenly distributed in all products yet, that's where engineer…”
Swyx: Application founders must pick vertical problems over technical AI solutions
“Actually, don't pick the solution, pick the problem. And if you're like, okay, I'm like, whatever it is in AI, whatever the hot thing is, whatever the new trend is, whatever the new model is, I will be the AI guy for dentists, or for lawyers, or for finance pe…”
Swyx: A full world model must be recursive and self-aware
“A full world model must also be recursive, meaning that the participant in the world model must also be aware that they have a world model. Which is like this whole recursive thing down the line. But yes, and that the world model can be wrong, and that they ne…”
Swyx: TypeScript could win the AI engineering ecosystem over Python
“I think it could be that TypeScript is going to win AI engineering, and that's something I haven't anticipated or seen, or I don't even know how to sort of Position around this, because I think it really does mean a lot of different things for what frameworks …”
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…”
Swix: Frontier model context windows will not reach billion-token scales
“We took, you know, three years to go from 100 K context to one million context in every frontier model, but we're not going to a billion, you're not going to a trillion with context graphs, context lengths.”
Swix: Vibe-Coding Tools as a Manager Creates Messes for Employees
“I think one failure mode of the post-technical, you know, like manager, which you are, which I am, is that you think you, oh, AI can do it. And then you like, you know, you vibe code something, you throw it to your employees, and then you expect that to just p…”
Swix: Claude Cowork is the closest feeling to AGI this year
“It is the closest I've felt to AGI this year.”
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…”
Swyx: AI agents should maintain personal wikis for persistent memory
“I think like almost like every agent should have its own wiki that it's updating and that's, Persistent memory. That is a very weak knowledge graph. And you could strengthen it if you want more structure, but you may not need it.”