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 930 resolved? a statement only gets an assessment when the public record can support or contradict it. opinions and what-ifs never can, and 8 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
Helberg: DeepSeek lied about compute capacity and distilled ChatGPT weights
“The main takeaways of how DeepSeq achieved the performance its performance was Incremental efficiency gains. They lied about their compute capacity because they have a billion dollar cluster, and they distilled ChatGPT's model weights.”
Prince: Prominent AI companies use state-hacker tactics to bypass scraping blocks
“In fact, some really prominent companies That do some very, very, very shady things where they basically say, absolutely, we follow the rules of the road, but when push comes to serve, if it turns out they're blocked, then all of a sudden they're doing a bunch…”
Caruso: Mayor Bass cut LAFD budget, leaving equipment mothballed in lots
“The city didn't even have enough equipment because they've underfunded the fire department for so long, and Mayor Bass cut even more out this last budget cycle That they had equipment mothballed in parking lots.”
Stability AI Has More Compute Available Than Microsoft or Google
“The reality is we have more compute available to us than Microsoft or Google.”
Tan: President Trump asked him to resign over conflicts of interest
“Well, the most surprising that I don't learn from my previous job or even training is one day early morning, President Trump asking me to resign and a conflict of interest, and there's no exceptions.”
Patel: Anthropic has surpassed OpenAI in API revenue
“Anthropic has eclipsed OpenAI and API revenue, and their API revenue is primarily not thinking. It's clawed for, but it's not in the thinking mode.”
Chen: Researchers Degrade AI Factuality Just to Boost LMSYS Rankings
“A lot of researchers, they'll tell us that their VPs make them focus on increasing their rank on LMSYS. And so I've had researchers explicitly tell me that they're okay with making their models worse. Add factuality works at following instructions as long as i…”
Valar Is the Only Private Company to Ever Produce Nuclear Power
“This is the fifth new nuclear device to make power in the United States since the year 2000. Fowler is the only private company founded since the discovery of nuclear fission that has ever made nuclear power.”
Gil: California is discussing an exit tax on departing residents by 2028
“And my sense is in 28, there's increasing talk about also trying to add a exit tax in California. So if you actually try and leave, they'll, they'll try and take a big chunk as sort of a penalty for that.”
Brown: Modern AI models can reason for weeks before plateauing
“What we're seeing today with the modern models is that 5.5 and other models can think for, if you scaffold them reasonably well, can think for weeks even before having performance plateau on some of these benchmarks.”
Tan: Jensen Huang's $5B Intel investment is now worth $25B+
“Secondly, very happy that Jensen Huang, my old time friend, he also put five billion in investing and support me. And I'm glad I at least do some good work. It's five billion become twenty five billion now and or more.”
Helberg: Belt and Road countries frequently default, yielding equity to China
“It is Chinese workers, Chinese companies, you know, and a lot of the times it's Chinese equity because the debt converts to equity if, you know, the host country defaults. They do a lot of the times.”
Huang: China contributes more to open source than any other country
“No country in the world contributes more to open source than China.”
Liemandt: Alpha School yields top 1% academic performance in two hours daily
“And so we built a learning engine that teaches kids 10 times faster. It's all based on the principles of learning science. And so then now we can commit to kids. You just need to do two hours a day and you're going to crush your academics. So top one percent p…”
Liemandt: Private school A-students test up to three grades behind
“Your student academically
On our standard, on standardized tests that we use are somewhere between one grade level ahead.
If they had a straight A transcript transferring in to us, one level, one grade level ahead to three grade levels behind.
And if they're a…”
Liemandt: Mastering a grade-level subject takes 20 to 30 hours
“The average grade level subject combo, so fourth grade math, takes between 20 and 30 hours
To master.
That's it.”
Nadler: OpenEvidence is used daily by 40% of US doctors
“We're used by 40% of doctors in the United States daily on average. It's about 20 times as much usage as the next thing that could be described as a clinical decision support platform.”
Patel: Microsoft dropped Stargate for OpenAI due to slow execution
“Microsoft is not building
Stargate for OpenAI, right? It's because it would have just been too slow, and they're doing it the lame and old way.”
Patel: Huawei Obtains TSMC Wafers via Shell Companies
“Huawei is still getting you know, wafers in Taiwan from TSMC through, like, shell companies, right?”
Prince: OpenAI and Anthropic Refer Traffic 750x and 30,000x Less Than Google
“In the case of someone like OpenAI, it's 750 times harder than it was with the Google of old. With the case of Anthropic, it's 30,000 times harder than the content of old”
Krishnan: Biden Diffusion Rule Effectively Banned US GPU Exports to Allies
“Under the Biden era, there was something called the Biden diffusion rule, which basically was a 200 page document, which basically made it illegal for America to export GPUs. It was really hard for you know, if you're Jensen or if you're Lisa Su to really kind…”
Dent: Chai-2 hits nearly 20% wet-lab antibody binding success
“We have about a two week validation cycle in the lab, and two weeks later, we see that roughly close to 20% of these antibodies actually bind their targets in the intended way.”
Mann: Competitors ran 'code reds' to match Claude in coding and failed
“And I know that other companies have had like code reds for trying to catch up in coding capabilities for quite a while and have not been able to do it.”
Mann: Opus 4 triggered ASL-3 safety protocols due to biological threat capabilities
“And so one of the reasons that our most recent model, Opus IV, is classified as ASL III. Is because it did have significant uplift relative to a Google search.”
Caruso: LA reservoir was drained for repairs despite high fire risk
“There's a whole reservoir that was empty. That was drained because they wanted to have repairs on it. Well, we had a fire in Malibu, 10 minutes from the Palisades fire, three weeks before. That's probably a pretty good sign that we needed to be prepared, and w…”
Caruso: Palisades hydrant failure was caused by empty reservoir, not broken pipes
“The hydrants are empty. We're not getting water. It had nothing to do with broken pipes. It had everything to do with the reservoirs not being filled. Everything up there is gravity flow. So those reservoirs draining are coming into the hydrants, and there's p…”
Wang: AI gets no positive transfer across modalities like video to text
“My understanding there's no positive transfer from learning in one modality to other modalities. So like training off of a bunch of video doesn't really help you that much with your text problems and vice versa.”
Wang: No strong evidence that video creates useful AI world models
“I don't think there's strong scientific evidence of that yet. Maybe there will be eventually.”
Mansour: Kalshi is the sole legal, regulated US election trading exchange
“So we won this month, and the cool thing now is, like, it's the first time in a hundred years that betting on the election, trading on the election is actually legal in the U.S., It's finally back to actually being legal, and this time it's actually regulated,…”
Mansour: Kalshi predicts daily weather better than most stations
“Like we can predict daily weather more accurately than most weather stations.”
Khan: FTC Finalized Rule Banning Most Employment Non-Compete Clauses
“A few months ago, the FTC finalized a rule that would ban non-compete clauses in the vast majority of employment contracts.”
Alexandr Wang: Cross-modality training produces no positive transfer between video and text.
“I think the main thing, fundamentally, is I think there's very limited generality that we get from these models and even for multimodality, for example my understanding there's no positive transfer from learning in one modality to other modalities. So like tra…”
Mensch: LLMs are not marginally more dangerous than web search
“Nothing is showing that LLM is actually marginally better than a search engine to find knowledge on, on topics that would enable bad use.”
Mensch: AI bioweapon risk narrative relies on unscientific circular citations
“No scientific studies in proper form was published, but then policy papers started to cite Non-scientific papers arguing that these were scientific evidences that the bioweapon narrative was actually true. And then policy papers started to cite the other polic…”
Mensch: Zero evidence AI development is headed toward a singularity
“If we can make a model which is growingly intelligent, then maybe you're at a singularity level. There's no evidence whatsoever that we are on the way of doing that, of making that happen.”
Gil: SARS leaked from laboratories four times in first two years
“Because SARS, they both snuffed out with the original form, but then it leaked four times in the first two years after it was cultured in a lab. It kept leaking from what eventually became the Wuhan Institute of Biology when they moved it from Beijing to Wuhan…”
Drone delivery competitors have spent billions without serving a single customer
“There are other companies in, you know, the drone delivery industry that I think have taken more of a moonshot approach and they've spent billions of dollars and have not served a single customer yet.”
Haas: Nvidia, Amazon, Microsoft, and Google Backed Arm's Physical Chip Strategy
“So whether it was Nvidia, Amazon, Microsoft, Google, All people who build ARM-based server chips, they were all on board, and I think the ultimate proof point was when we announced the product last March, we had Jensen, we had Ronnie Boker, we had Amin, we had…”
Hodak: Science is the first to restore form vision in blind patients
“Nobody had previously ever been able to restore a form vision image in the mind's eye of a blind patient in this way.”
Hodak: Science aligns animal brain recordings with AI model representations
“We use that constructively at science like that. So we know that that is. That is true because we can get alignments between animal brain neural recordings and AI model internal representations.”
Taylor: Nuclear energy causes fewer deaths per unit of power than solar
“Nuclear energy is the safest form of energy, empirically. If you look at the amount of power generated versus the amount of deaths it's even safer than solar, which is very strange, because how does someone die from solar? But actually, most solar is installed…”
Valar Built a $5M Nuclear Reactor Component In-House For $400K
“And we were quoted about five million dollars and we were told it would take about two and a half years to build. And so we said, well, all right, maybe I'd be willing to pay five million dollars, but I'm certainly not going to wait two and a half years. And s…”
Guo: Researchers have predicted ASI is 18 months away for five years
“The data I have is that a number of very smart and even very self-aware research scientists have felt that, you know, there was some knee in the curve on recursive self-improvement or ASI, 18 months away, every 18 months for the last five years.”
Gil: AI labs are slowing researcher hiring due to compute costs
“I know some labs have slowed down on their hiring of researchers unless they're above a very, very high bar because the cost isn't the research or it's the compute associated with the person.”
Fogel: Booking has grown faster than Airbnb in alternative rentals for 5 years
“For the last five years, we've grown faster than Airbnb in the alternative accommodation area. We do like to like. We've been growing faster the last five years.”
OpenAI internal model reportedly disproved the Erdős unit distance conjecture
“We used an internal model at OpenAI a few weeks ago to disprove the unit Erdos unit distance conjecture.”
Brown: AI cannot invent novel algorithms better than existing research
“Go ahead and like look at all the published work and synthesize that and then try to come up with something novel and it's not able to do it. And I can give it a lot of time and it's still not able to do it.”
Tan: AI inference is shifting CPU-to-GPU ratios toward 1:4 or 1:1
“Right now, the authentic AI and inference, CPU become, you know, highly in demand. And so, you know, versus one to eight in the training CPU to GPU, now I can see one to four, maybe one to one, and I'm delighted CPU become important.”