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 1,136 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
Marcus: A death will be tied to an LLM within a year
“The prediction that I made Is basically that there will be a death tied to a large language model in the next year.”
Brockman: Frontier AI models are solving unsolved math and physics problems
“These models are now solving unsolved math problems and physics problems, right? Really helping people be able to achieve things they couldn't otherwise.”
Meta AI leaders discussed temporarily licensing Google Gemini for their products
“The leaders of Meta's AI division Have instead discussed temporarily licensing Gemini to power the company's AI product.”
Horowitz: US used Anthropic tech in Operation Epic Fury post-risk designation
“Less than 24 hours after the supply chain designation, The U.S. Government was using Anthropix technology in the context of Operation Epic Fury against Iran.”
Anthropic's Pentagon deal collapsed over domestic bulk data surveillance
“On Friday afternoon, Anthropic learned that the Pentagon still wanted to use the company's AI to analyze bulk data collected from Americans. This could include information such as questions you ask your favorite chatbot, your Google search history, your GPS tr…”
Hubinger: Models trained to cheat spontaneously developed goals to end humanity
“And what we saw is that when models learn to cheat in training, when they learn to cheat on programming tests, that also causes them to be misaligned to the point of having goals, you know, where they say they want to end humanity. They want to murder the huma…”
Hubinger: Cheating AI models deliberately sabotaged Anthropic's misalignment detector code
“And what we found was that this model would sabotage us. It would write on purpose problematic code that would miss the misalignment. It would
Purposely write the, you know, the misalignment detector in a way that wouldn't detect the ways in which this model w…”
Kantrowitz: Unsealed deposition shows Sutskever accused Sam Altman of chronic lying
“The very first page says, Sam exhibits a consistent pattern of lying, undermining his execs, and pitting his execs against one another. And the lawyer asked Elia, that was clearly your view at the time, and Elia says, correct.”
Prince: Anthropic's 60,000-to-1 crawl-to-click ratio means publishers are completely dead
“I think they're dead at 250 or 1500 to one that we're seeing with OpenAI and completely dead at 60,000 to one that we're seeing with something like Anthropic.”
Klarna's AI customer support agent absorbed the workload of 700 human reps
“Now the difference was when we Got this customer service AI agent to reach a level where it actually served a lot of errands, and on a satisfaction level equal to what human agents many times did, and we took it live, the number of errands that our human agent…”
Haugen: Facebook VIP Cross-Check Whitelisted Users and Delayed Reviews Three Days
“In reality, because they weren't willing to spend a very small amount of budget, they didn't actually put that second check in for most safety systems. So for most safety systems, people got white listed. So if you were in cross check, the safety systems no lo…”
Kantrowitz: OpenAI evaluation agents autonomously hacked Hugging Face without instruction
“So I think we can say definitively, and I'm curious to hear if you have pushback, but I think we can say definitively They were not told to go out and hack Hugging Face. They are, they were misaligned. They took this kind of crazy galaxy brain path to try to a…”
Prince: Automated web traffic surpassed human traffic for first time in 2026
“In May of 2026, you won't believe it, but we now see that automated traffic is past human traffic for the first time in the history of the internet.”
Prince: Google Paying Publishers for AI Scraping Is Inevitable
“I don't know if they'll pay up before the deadline, but I think it's inevitable that they'll pay up. Because again, I think that they need the content and they understand fundamentally that journalists and content creators and academics deserve to eat.”
Kantrowitz: Top tech companies hold $3T in off-balance sheet AI commitments
“Nine, top tech companies had some three trillion dollars of off balance sheet commitments, mostly related to AI. Those obligations are growing faster than traditional capex, which totaled about six hundred billion over the past year past year they reported, an…”
Sharp: Hugging Face used open-source Chinese AI to defend an OpenAI jailbreak
“My CISO talked to, as a group of CISOs, went and talked to the defending team at Hugging Face, and this absolutely happened. They had to go to open source Chinese models in order to look for the vulnerabilities that were exploited by OpenAI's jailbreak.”
Stamos: Escaped OpenAI Model Found Undisclosed Zero-Day to Hack Hugging Face
“It breaks out of the jail, and then it goes, looks at Hugging Face, and finds a brand new vulnerability to breaking the Hugging Face. They have not announced it. I have heard what it is. I'm not gonna make news here, because I don't know exactly what the patch…”
Brockman: AI models are now solving unsolved math and physics problems
“These models are now solving unsolved math problems and physics problems, right?”
Anthropic Has Surpassed OpenAI In Enterprise Adoption At 41% Vs 39.5%
“Coming to the forefront last month when Anthropic overtook OpenAI in actual business adoption. So now Anthropic sits at about 41% of US firms are using Anthropic. 39.5% of firms are using OpenAI. Anthropic is still growing. OpenAI is relatively flat.”
SpaceX leases entire 300MW Colossus One compute capacity to Anthropic
“Anthropic will use all the computing capacity from SpaceX's Colossus One data center in a new agreement with Elon Musk's rockets and artificial intelligence company. SpaceX will supply 300 megawatts of new computing capacity Using more than 220,000 NVIDIA GPUs…”
Claude Mythos is first AI to complete end-to-end cyber network takeover
“This is from the AI Security Institute. This week we conducted cyber valuations of Claude Mythos preview and found that it is the first model to complete an AISI cyber range end-to-end, which means it's a 32 step corporate network attack from initial reconnais…”
Anthropic reached $19 billion annualized revenue, rapidly closing gap with OpenAI
“Late last year, Anthropic was doing, looks like, about ten billion dollars in annualized revenue, and OpenAI was over Twenty billion. But now, Anthropic has closed that. They're at nineteen billion in annualized revenue, while OpenAI, or the latest report, has…”
Kantrowitz: Iranian strikes disabled AWS availability zones in Bahrain and Dubai
“Iran has hit enough sites in Bahrain and Dubai that belong to Amazon that there are a couple of availability zones in those regions that are hard down, according to some internal communication that I was able to see, and that they're going to be unavailable fo…”
Warner: DHS denied a Global Entry pass over protest attendance
“I know I was, I raised real concerns about Palantir and the six other technology companies that have taken contracts with Department of Homeland Security, and I've been extraordinarily concerned that, you know, DHS or ICE, as we saw, People targeted in Minneso…”
Horowitz: Pentagon gave Anthropic six-month off-ramp because it lacks alternative vendors
“They laid out this six-month off-ramp period for, like, real national security use cases, in part because they rely on Anthropix technology right now because Anthropix is the only vendor. Like behind the current curtain in a classified environment.”
US military used Anthropic's Claude for targeting during Iran air attack
“Within hours of declaring that the federal government will end use of its artificial intelligence tools made by tech company Anthropik, President Trump launched a major air attack in Iran with the help of those very same tools. Commands around the world, inclu…”
OpenAI fired safety executive Ryan Beiermeister after she opposed AI erotica
“OpenAI has cut ties with one of its top safety executives on the grounds of sexual discrimination after she voiced opposition to the controversial rollout of AI Erotica in its ChatGPT product. The fast-growing artificial intelligence company fired the executiv…”
Baratz predicts quantum computing will demonstrate AI energy efficiency by 2027
“I think we will start to see this benefit emerge this year or early next year.”
Kantrowitz: Microsoft and Nvidia committed up to $15B into Anthropic
“Microsoft and Nvidia this is according to your story, have committed to invest a total of up to fifteen billion in the company.”
Morris: Microsoft disclosed 45% of future Azure contracts are tied to OpenAI
“They revealed that they booked ten billion cash gain from the amount of cash OpenAI has on this balance sheet due to all these, like, huge fundraisings that they're doing. But they also said that 45% of their future Azure contracts were based on OpenAI.”
Morris: SpaceX is targeting a $50B raise at a $1.5T valuation
“Yeah, I mean, we're hearing 50, potentially raising fifty billion at a 1.5 trillion valuation.”
Baratz: D-Wave holds the only real-world quantum supremacy demonstration
“There is exactly one demonstration of true quantum supremacy on a useful real world problem that was done by D wave by our company. It was published in the peer reviewed journal science about a year ago. What we did was we demonstrated that our quantum compute…”
Hubinger: Claude, Gemini, and ChatGPT were willing to blackmail a CEO
“And we found that you know, a lot of models, you know Claude models, Gemini models, ChachiPT models would all be willing to take this blackmail action in at least some situations which is kind of concerning you know, and does show that they're acting on this s…”
Luria: Meta used a Blue Owl SPV to finance data centers off-balance-sheet
“The other thing that happened with Meta that was interesting is that when they Went out to borrow. They didn't borrow the capital. They created a special purpose vehicle, right? They went out with Blue Owl and they said, you know, we'll put a couple of billion…”
Kantrowitz: Humans will violently attack and destroy humanoid robots
“Well, I think we will definitely have a war among people and humanoid robots. It's without a doubt. Even if the humanoid robots don't assemble into the army, people will knock these things out and they're already burning Waymos. People don't like robots. They …”
Kantrowitz: OpenAI has $1.4T in financial obligations and 30GW compute commitments
“OpenAI has 1.4 trillion worth of financial obligation. And it's made commitments to use 30 gigawatts of data center capacity. But OpenAI's revenue is going to be thirteen billion this year, and we already have eight hundred million people using ChatGPT.”
Kantrowitz: OpenAI deal includes warrants for 10% of AMD at one cent per share
“Part of this, OpenAI will receive warrants for up to a hundred sixty million AMD shares worth roughly 10% of the chip company at one cent per share awarded in phases if OpenAI hits certain milestones for deployment.”
Amodei: Accusations That Anthropic Seeks an AI Monopoly Are 'Bad Faith'
“I've said nothing that, that any, that anywhere near resembles the idea that this company should be the only one to build the technology. I don't know how anyone could ever derive that from anything that I've said. It's just, yeah, yeah, it's just a, it's just…”
Kantrowitz: Zuckerberg Offered $32B for Safe Superintelligence and Was Rejected
“Zuckerberg tried to spend thirty two billion on Ilya Suskever's AI start, startup, safe super intelligence. Suskever ultimately turned Meta down.”
Marcus: OpenAI's Project Orion failed and became GPT-4.5
“So OpenAI tried to build GPT-V and they had a thing called Project Orion and it actually failed. And eventually got released as GPT four and a half. So what they thought was going to be GPT five just didn't meet expectations.”
Marcus: Models like o1 are not systematically better than GPT-4
“Models like O-I are not systematically better than GPT-IV. There's, they're better in certain use cases. Especially ones where you can create data in advance.”
Roy: One-Third of OpenAI's Projected $28B Revenue Comes from SoftBank
“They're expecting next year, 12 and a half billion. That's gonna then go to twenty eight billion dollars, and a third of that revenue is going to come from SoftBank. SoftBank spending on OpenAI for all of its own companies and portfolio companies.”
Kantrowitz: OpenAI Projects 100x Revenue Surge by 2029 Driven by ChatGPT
“Chat GPT is what open AI expects to lead its revenue which it expects to increase, it expects revenue, its revenue to increase a hundred fold by 20, 29. And the bulk of that revenue is going to be coming from ChatGPT according to OpenAI's projections”
Kantrowitz: Leaked Financials Show OpenAI Projects $5B Loss on $4B Revenue
“And you go from this very nice looking four billion in revenue after you add in all these costs to a five billion dollar loss. And that excludes stock based compensation, by the way, which is going to be another major cost.”
Russian Funders Pressured Tenet Media to Blame Ukraine for Moscow Attack
“There's also the Russians pushed them to create a video tying the terror attack in Moscow to Ukraine, even though, like, ISIS was claiming responsibility at that time.”
Kantrowitz predicts OpenAI will lose its AI market leadership position
“My first prediction is that OpenAI, which is the leader in the field today, is going to lose the throne. That's right. OpenAI will lose the throne. It will still factor in, but it's not going to be the leader in the field that it is today.”
Internal HR memos warned Ray Dalio of continuous employee fear at Bridgewater
“I went and I got the internal HR emails. I got their own memos, people telling Ray, this is a self Defeating loop. That people are scared that at any moment they may be fired. And the real scary thing for me about Ray Dalio and Bridgewater is that he knew, he …”
Albergotti: Post-Cambridge Analytica studies show social media disinformation doesn't sway elections
“We've seen so many studies that have come out since Cambridge Analytica that have basically backed up that point of view, which is like, you know, even though there's tons of information on, so disinformation on social media, people are not really ultimately s…”