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 485 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
Patel: AI is underearning the economic value it creates by a significant margin
“AI is under earning the value that it's producing in the world, right? By a significant margin already today.”
Patel: Opus 4.5 on Claude Code permanently changes how people work
“Opus 4.5 on Claude code is a new moment where the way you work has forever changed.”
Text diffusion models will not replace autoregressive Transformers at state-of-the-art
“So it is a interesting direction to go into these diffusion, diffusion models as alternative to the auto regressive transformers, but it is not I would say the replacement at the state of the art.”
Sovereign AI initiatives are mostly marketing and sovereignty washing
“I think it's more marketing than it is like a real policy, because at the end of the day, if you buy your stock from the U S and you're not an ally of the U S at some point, then they'll just switch it off. And so part of this is like sovereignty washing, I th…”
Schrittwieser: Wider AI ecosystem may face bubble while frontier labs thrive
“There may simultaneously be like some sort of bubble in, you know, the wider ecosystem, while at the same time, the frontier labs on a very solid trajectory, having a lot of revenue, making a lot of money.”
Tworek: OpenAI's o1 was mostly a tech demo for solving puzzles
“O-one like, to be perfectly honest, it was really mostly good at solving puzzles and like maybe a few kind of thinking problems here and there, but it wasn't like, it wasn't a very useful model. It was almost more like a technology demonstration.”
Tworek: The landmark 2012 ImageNet AI results were not that significant
“From my perspective, and again, this is just how my brain works, that the 20 12 ImageNet results, like, weren't that significant.”
Douglas: Claude 3.5 Sonnet drove product-market fit for code editor Cursor
“In many ways, this model is what caused PMF for Cursor. Cursor took off like a rocket, right, with because they were in the right place, and they were able to capitalize on that model as offering a coding experience that didn't previously exist.”
Douglas: LLM pipelines are two and a half years of desperate effort
“When I look at an LLM training pipeline, it is two and a half years of best effort, last minute, desperate effort. And there's just so much room to go on every part of it.”
Valenzuela: Runway's true moat is its shipping infrastructure, not any individual model
“And I think that for me is the most valuable part of runway. It's not a model that we put out because models will like Completely like change every single, every now and then is the organizational knowledge and the infrastructure that it takes to ship a model …”
Cherny rates Claude Code 10/10 for codebase research, 6/10 for writing
“I feel like if you look at answering questions about the code base and kind of code base research, I think it's like 10 out of 10 good. It's as good as it can get. When it comes to writing code, it's maybe like a six out of 10. It's pretty good. It won't get e…”
Rauch: The total addressable market for v0 is everyone with a computer
“When I think about the coding IDEs, You still have to be a coder, meaning you know how to download an ID, you have to, you know how to set up, you know, the dev server, the tooling, the SDKs, the libraries, whereas a product like vZero, the TAM is everyone wit…”
Canva's proprietary design data uniquely positions it to build foundation models
“So maybe that advice of kind of like, don't be in the foundation, foundational model Space is good general advice, but we think that we are in a quite unique position in terms of the data that we have to pursue that particular avenue.”
Chatbots are like calculators, inadequate for serious enterprise knowledge work
“Chatbots are, in my eyes, like the TI-eighty-four, or like the HP-twelve-c, like a calculator. It's a one-off, you know, I'm gonna go and put in my equation and then get a response. Nobody does their taxes in a calculator. Nobody, you know, computes a DCF or, …”
The San Francisco artificial intelligence ecosystem suffers from strong groupthink
“I think not a lot of people are saying it, but I think like SF has a really big group think.”
Howard: There was no technological breakthrough 'DeepSeek moment'
“For me, there was no technology DeepSeek moment.”
Jeremy Howard: Academic approaches to predictive modeling are far less successful than practical experience
“Much to my surprise, the academic approach is, was way less successful.”
Sutter Hill's Mike Spicer funds companies despite overwhelming failure evidence
“Which is, by the way, a testament to Mike Spicer at Sutter Hill because he is, he has so much conviction in his companies that despite overwhelming evidence that the company is a total failure, he keeps investing in it.”
Mike Palmer: Forcing SQL concepts into natural language is engineering hubris
“So the idea that I'm supposed to adopt a SQL semantic concept into my natural language is the ultimate hubris of engineering.”
Ramaswamy: DeepSeek is one or two steps behind xAI in model training
“If you were to compare them to XAI, I would say they are definitely one or two steps behind in terms of their ability to come from nothing and train a world-class foundation model.”
Chollet: OpenAI o3 is the most advanced test-time adaptation model
“And OSTRI best I can tell is the most advanced the most successful test and adaptation model out there at this time.”
Misra: Video generation is probably the most dangerous AI unlock
“Because I think of all the possible sort of AI unlocks, like even compared to LLMs and stuff, there's probably nothing more dangerous than video. Generation in a way, right?”
AI venture investing is rebuilding another 2021-style bubble
“Half of your time, you basically untangle the consequences of the excess period of twenty-twenty-one, where you know, you think of helping, how to help company get fit and you know, cut burn, extend runway, be efficient, all the things. That's like one part of…”
Delangue: AI regulation should address present risks, not doomsday scenarios
“In general, for me, I'm in favor of the regulations that are creating more competition, more opportunities for everyone to build with AI, and tackling some of the end user risks of AI today, like the risk of misinformation, the risk of biases. Instead of focus…”
Pomel: Many big tech AI assistants risk being too much like Clippy
“The risk right now is that a lot of the assistants we're seeing from the big companies are too close to Clippy.”
Glyman: Silicon Valley workers are mercenaries averaging 12-14 month tenures
“It's not like being in the valley where people are mercenaries. You know, I think the average career tenure is, you know, 1214 months.”
Adding sequential LLM calls or filters to catch errors fails in production
“It's both of those things, and I think people are addressing error today by adding more calls to the model of filtering. Out the requests. And I think I don't think that'll work for serious production use cases.”
Jeremy Kahn: AI regulation should focus on present risks over existential fear
“I do not think that that fear should be the basis on which all AI regulation is based. And I do not think it's should crowd out efforts to police risks that are very real and here now around racial bias, around disinformation around job loss.”
Kahn: Tech companies cannot be trusted to self-regulate AI
“I do not think that companies can be trusted really to do this completely on their own. I think there's just too much profit motive to take actions that ultimately will be kind of detrimental to us from a societal level or personal level.”
Starting later than Snowflake and Databricks was VAST Data's biggest advantage
“And so I think our The biggest advantage versus those other companies that you mentioned is that we started late, and that allowed us to see this deep learning problem in front of our eyes versus machine learning, again, a lot smaller, a lot slower, and it als…”
Traynor: Customer service is extremely vulnerable in AI's 'kill zone'
“I think genuinely customer service is very much in like the kill zone of AI or whatever. I think it's really you know, it's a very vulnerable space if you don't move with the times.”
Duderstadt: Unequal access from market consolidation is the top AI risk
“What I would say the number one risk of this technology is, which is the sort of unequal access element to it. Like the version of the world that I fear most is like, there's two or three mega companies that like lock everything down in like the early days.”
Huber: Current SOTA LLMs Lack Reliability for Multi-Agent Workflows
“Now, of course, for those of you that have actually played with technology, I think it's questionable whether the current state of the art Language models, embedding models, et cetera, will give you the reliability you want from, ah, you know, agents working t…”
Mooney: A large part of tech wage gap stems from negotiation reluctance
“A large part of the reason the wage gap exists is that like women and minorities in technology are not not necessarily willing or confident to negotiate.”
Katz recommends Datadog for hosted APM over stitching custom solutions
“If you are looking to deploy a hosted APM solution, and you asked me objectively, my recommendation, I'd say use Datadog. I'd say that the product, the maturity, the experience is, is considerably better than trying to stitch together something using ClickHous…”
Pinterest visual search beats Google and Bing in head-to-head tests
“You'll find that Pinterest is usually the one that comes out of top.”
Apache Spark failed to effectively shrink down to single-node scale
“One of the things that you find with things like Spark is that they really failed to shrink down and, Effectively do computing at the single node scale.”
Pesenti in 2020: Large neural chatbots are not ready for consumer products
“It's not ready for prime consumption, let's be clear. You know, like these bots have lots of interesting side effects. You know, you cannot really control what they're gonna say. But they are quite entertaining, quite engaging, and a lot more human-like than y…”
Public markets rightly penalized Cloudera for lagging on cloud software transition
“I believe that Cloudera could be a lot bigger and a lot more successful but the market has been penalizing Cloudera, rightly so, for not capturing the cloud's components as sooner.”
Delangue: NLP is the most important field of machine learning
“I believe that NLP today is the most important field of machine learning.”
Volpi: VC industry and Index Ventures have failed on diversity
“I think the industry at large, and our firm included in this, has not done a good job at all Funding against diversity or including diversity in our own firms?”
Pedro Domingos: Very little of current machine learning capability has been deployed
“With the existing machine learning technology, of what we can do with that, how much have we done? Very little so far. So there's enormous amount for, enormous scope to do things there. Not just in finance, but in a lot of other fields, right?”
Marcus: Deep learning fails at reading comprehension and misses the point
“I think that deep learning doesn't just fail on reading. It completely, entirely misses the point.”
Worrying over $10M versus $20M early-stage startup valuations is myopic
“For me to worry about the difference between 10,000,020 million valuations seems like such a myopic view of the fact that technology is working, and it's disruptive, and the market is like basically the globe.”
The NYT cannot build a viable business on third-party tech platforms
“We don't think it's a good delivery vehicle for our product, and we also don't think we can build a business in any of the platforms, really.”
Dixon: Betting against Ethereum means betting against top developer talent
“If you want to bet against Ethereum as an example, like you're effectively betting against, you know, I don't know, like a lot of really, really smart engineers, and I just, I think it's a, You know, I'll take the other side of the bed.”