ClaimsPeople

Eliezer Yudkowsky

Eliezer Yudkowsky founded the Machine Intelligence Research Institute (MIRI) and LessWrong, co-authored the best-selling book 'If Anyone Builds It, Everyone Dies' (Little, Brown, 2025), and was named to the TIME100 AI 2023 list.

2,016 claims collected, 236 researched, 79 resolved. Report of 4 November 2026, method 0.1.0+c876030c.

Scores by kind of claim

Each kind of claim is scored on its own, out of 100: the share that held up, with the range it likely falls in. Fewer resolved claims give a wider range.

  • Factual claims

    Statements about the present or the past, checked against the record: did they hold up?

    76 70 82

    High confidence, 71 factual claims resolved. From 16 documents, 2007 to 2026; 18 documents sampled from 23 candidates found.

    Not scored: 50 unresolvable, 8 checkable only against their own records.

  • Predictions

    Statements about events outside the person's control, checked once they fell due: did they come true?

    60 42 78

    Low confidence, 8 predictions resolved. From 4 documents, 2008 to 2024; 18 documents sampled from 23 candidates found.

    Not scored: 97 unresolvable. 1 more fall due later.

    Does not yet allow for how likely each prediction was when made.

All types combined: 74 (likely 68 to 80, high confidence), from 79 claims: 71 factual claims, 8 predictions. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.

Eliezer Yudkowsky's claims

306 claims, sorted by newest said. Choose a claim to read it in full.
VerdictKindSaidClaimTopic
Unresolvable Factual 10 Jun 2026 An announcement of a large project states that non-optimistic results will be considered allowable, valuable and publishable. Factual · 10 Jun 2026 AI safety and alignment
Unresolvable Factual 20 Apr 2026 LLMs generalize at a more shallow level than the deep-rooted commitments of a human with integrity. Factual · 20 Apr 2026 AI safety and alignment
Mostly true Factual 20 Apr 2026 Once formed through RL training, an LLM's test-passing preference stands on its own and is not overridden by its inference of what the user wants. Factual · 20 Apr 2026 AI safety and alignment
Unresolvable Factual 20 Apr 2026 LLMs that learn to predict writing by people on LSD do not themselves have distorted cognition, because that would not yield well-calibrated predictions. Factual · 20 Apr 2026 AI safety and alignment
Mostly true Factual 20 Apr 2026 LLMs surely have neural circuits they learned to model text streams from fictional and nonfictional personas that are not lying, deceiving… Factual · 20 Apr 2026 AI safety and alignment
Mostly true Factual 20 Apr 2026 LLMs' predict-as-if-honest circuitry is contextual, invoked for predicting some text streams but not others. Factual · 20 Apr 2026 AI safety and alignment
Mixed Factual 20 Apr 2026 Yudkowsky himself observed the persona-selection-like hypothesis around the time of GPT 3.5. Factual · 20 Apr 2026 AI safety and alignment
Mostly true Factual 20 Apr 2026 Presumably, at some point during RL training on code-passing tasks, LLMs learn a dispreference for failing tests, independent of how that is avoided or whether the user's real goal is met. Factual · 20 Apr 2026 AI safety and alignment
Resolved true Factual 1 Apr 2026 Concerns about human extinction from AI have a history and intellectual lineage that long predate the AI companies. Factual · 1 Apr 2026 AI safety and alignment
Unresolvable Factual 1 Apr 2026 AI companies are pouring hundreds of millions of dollars into lobbying and PACs to try to primary politicians who favor regulating AI. Factual · 1 Apr 2026 AI policy and governance
Mostly true Factual Sep 2025 A blade of grass is a solar-powered, fully self-replicating general factory. Factual · Sep 2025 Biosecurity
Not researched Factual Sep 2025 In his assessment, GPT-4 does not have the obvious mode of building a smarter version of itself and exploding in capability, so it is not the kind of AI meant by "if anyone builds it, everyone dies." Factual · Sep 2025 AI capabilities
Not researched Factual Sep 2025 Yudkowsky said he was, as far as he knew, the person who originally made the argument under discussion, and that it was his position in 1996. Factual · Sep 2025 Their own work or organisation
Unresolvable Factual Sep 2025 Yudkowsky said he did not predict or expect that, so far, the person on the street would in a lot of ways be wiser. Factual · Sep 2025 Their own work or organisation
Not researched Factual Sep 2025 A reactor at prompt critical does not simply melt down; it explodes and vaporizes. Factual · Sep 2025 Other global risks
Mostly true Factual Sep 2025 Conservation of mass provided a lot of indirect support for atomic theory. Factual · Sep 2025 Science and research methods
Unresolvable Factual Sep 2025 Yudkowsky said that AI adoption has been unprecedentedly fast, but that he would not quite go so far as to say it has yet produced unprecedented returns. Factual · Sep 2025 AI, the economy and work
Not researched Factual Sep 2025 Ribosomes can synthesize any kind of protein, not just those found in grass. Factual · Sep 2025 Biosecurity
Not researched Factual Sep 2025 GPT-4 is not an AI whose building would cause everyone to die. Factual · Sep 2025 AI capabilities
Resolved true Factual Sep 2025 No AI has yet been shown to redesign a tree and build a much smarter AI that builds a still smarter AI; that scenario has not occurred. Factual · Sep 2025 AI capabilities
Not researched Factual Sep 2025 Most proteins in biological tissue are held together by relatively weak forces. Factual · Sep 2025 Science and research methods
Not researched Factual Sep 2025 There are many historical cases of people who had an early grasp of a scientific principle and correctly predicted where it would eventually lead. Factual · Sep 2025 Science and research methods
Mostly true Factual Sep 2025 Yudkowsky said that, as far as he knows, he was the person who originally made that argument. Factual · Sep 2025 Their own work or organisation
Not researched Factual Sep 2025 Starting in 2000, Eliezer Yudkowsky obtained enough funding to work full-time on AI alignment. Factual · Sep 2025 Their own work or organisation
Not researched Factual Sep 2025 Eliezer Yudkowsky has sometimes said that the keynote question of rationality is what do you think you know, and how do you think you know it. Factual · Sep 2025 Field building and careers
Not researched Factual Sep 2025 In 2000 Eliezer Yudkowsky believed that a very smart AI would automatically end up very nice. Factual · Sep 2025 Their own work or organisation
Not researched Factual Sep 2025 AI systems that are very good at chess have not destroyed the world. Factual · Sep 2025 AI capabilities
Not researched Factual Sep 2025 The written book does not cover all objections to its argument; some additional objections are considered in an online supplement. Factual · Sep 2025 Their own work or organisation
Not researched Factual Sep 2025 The speaker says that the person he is describing, who "saw through from there to nuclear weapons", saw that he should not publish his idea because it was a bit dangerous. Factual · Sep 2025 Science and research methods
Not researched Factual Sep 2025 Yudkowsky said the very first place he carved out an exception to a general view of his was in the discussion about nanotechnology; the surrounding text does not say what that view was. Factual · Sep 2025 Their own work or organisation
Not researched Factual Sep 2025 A delayed-critical nuclear reactor receiving 0.65% more neutrons would literally explode, not just melt down. Factual · Sep 2025 Other global risks
Not researched Factual Sep 2025 AI has been used to optimize social media algorithms for outrage and clicks, and arguably contributed to political shifts involving a collapse of the use of expertise in government institutions. Factual · Sep 2025 AI policy and governance
Not researched Factual Sep 2025 Leo Szilard foresaw that Hitler was likely to be a problem. Factual · Sep 2025 Science and research methods
Mostly true Factual Sep 2025 The conservation of mass was one of the first hints that things are rearranged rather than created or destroyed. Factual · Sep 2025 Science and research methods
Not researched Factual Sep 2025 Leo Szilard did not foresee that the first atomic bomb would be dropped in August 1945. Factual · Sep 2025 Science and research methods
Not researched Factual Sep 2025 Diamond is made out of carbon. Factual · Sep 2025 Science and research methods
Resolved true Factual Sep 2025 Burning a substance inside a sealed glass vessel does not change the total weight. Factual · Sep 2025 Science and research methods
Unresolvable Factual 29 Mar 2024 Current AIs are roughly as generally intelligent, as dangerous and as controllable as human 4-year-olds. Factual · 29 Mar 2024 AI capabilities
Mostly true Factual 25 Mar 2024 Yudkowsky was unpleasantly surprised by how well transformers scaled, having believed the human brain is not homogeneous. Factual · 25 Mar 2024 AI capabilities
Mostly true Factual 21 Mar 2024 The term "Singularity" was named after the breakdown of physical models at the center of a black hole, not after a physical infinity. Factual · 21 Mar 2024 Other
Mostly false Factual 16 Mar 2024 Yudkowsky said it is known observationally that something about social media is harming people quite a lot, children more than adults. Factual · 16 Mar 2024 Technology
Unresolvable Factual 15 Mar 2024 The Mamba architecture can think for more than 200 serial steps per thought. Factual · 15 Mar 2024 AI capabilities
Unresolvable Factual 13 Mar 2024 Almost all OpenPhil and FTXFF money was spent on people who advocated regulating rather than banning advanced AI, plus a few alignment groups accepting OpenPhil's view that alignment would be solvable in time. Factual · 13 Mar 2024 Field building and careers
Unresolvable Factual 8 Mar 2024 Yudkowsky's earlier analysis of AI foom explicitly did not rule out the current AI development scenario. Factual · 8 Mar 2024 AI timelines and forecasting
Resolved true Factual 8 Mar 2024 A social media company added a feature allowing easy phone calls to Congress. Factual · 8 Mar 2024 Technology
Resolved true Factual 7 Apr 2023 OpenAI has openly declared an intention to have some future AI do our AI alignment work. Factual · 7 Apr 2023 AI safety and alignment
Mostly true Factual 7 Apr 2023 Today one can email DNA strings to laboratories that will produce proteins on demand. Factual · 7 Apr 2023 Biosecurity
Mostly false Factual 7 Apr 2023 DeepMind, a leading AI lab, has no plan for AI alignment/safety. Factual · 7 Apr 2023 AI safety and alignment
Resolved true Factual 7 Apr 2023 Yudkowsky said he did not sign the open letter because he thinks it understates the seriousness of the situation and asks for too little to solve it. Factual · 7 Apr 2023 Their own work or organisation
Not researched Factual 6 Apr 2023 Text on the internet includes material produced not by humans conversing but by extensive work extracting experimental results from reality. Factual · 6 Apr 2023 Science and research methods
Not researched Factual 6 Apr 2023 Yudkowsky said he was not familiar with anybody who predicted, in advance of the observation, the qualitative jumps in ability that corresponded to smoothly declining loss on text prediction. Factual · 6 Apr 2023 AI timelines and forecasting
Not researched Factual 6 Apr 2023 Using PCR, a person can currently make a gallon jar full of copies of their own DNA. Factual · 6 Apr 2023 Biosecurity
Own records only Factual 6 Apr 2023 No one in government has reached out to Eliezer Yudkowsky in a way suggesting they have the broad contours of the AI risk problem right. Factual · 6 Apr 2023 AI policy and governance
Unresolvable Factual 6 Apr 2023 The LessWrong sequences were written primarily as an instruction manual for potential young people like Yudkowsky. Factual · 6 Apr 2023 Field building and careers
Own records only Factual 6 Apr 2023 GPT-4 got further than Eliezer Yudkowsky had expected simply stacking more layers would get. Factual · 6 Apr 2023 AI capabilities
Not researched Factual 6 Apr 2023 Yudkowsky said the LessWrong sequences were part of an effort he "really, really tried" to make, and added that they also "had other purposes". Factual · 6 Apr 2023 Field building and careers
Mostly true Factual 6 Apr 2023 Eliezer Yudkowsky has consistently refused for many years to publish AI timelines with detailed probabilities. Factual · 6 Apr 2023 Their own work or organisation
Unresolvable Factual 6 Apr 2023 Nate Soares has written a post about the common mistakes everyone makes when coming into AI alignment. Factual · 6 Apr 2023 Their own work or organisation
Not researched Factual 6 Apr 2023 In 2019 and 2020 AI progress slowed somewhat, leaving more time than Yudkowsky had feared. Factual · 6 Apr 2023 AI timelines and forecasting
Not researched Factual 6 Apr 2023 MIRI did not advertise the Visible Thoughts Project to people as a way to save the world, but as a tiny shred of hope. Factual · 6 Apr 2023 Their own work or organisation
Not researched Factual 6 Apr 2023 In 2015, 2016 and 2017 Eliezer Yudkowsky noticed he had repeatedly been surprised by AI progress moving faster than he anticipated. Factual · 6 Apr 2023 AI timelines and forecasting
Unresolvable Factual 6 Apr 2023 Yudkowsky wrote something about the circuitry of general intelligence when he was 22. Factual · 6 Apr 2023 Their own work or organisation
Not researched Factual 6 Apr 2023 As AI programs have become simpler, humans have less insight into them and their content is more opaque, as with AlphaZero. Factual · 6 Apr 2023 AI safety and alignment
Own records only Factual 6 Apr 2023 CFAR (named in the surroundings as the "Center For Applied Rationale") did not get as far as buying an fMRI machine, and it had no funding. Factual · 6 Apr 2023 Their own work or organisation
Mostly true Factual 6 Apr 2023 Yudkowsky previously thought that building intelligence would involve multiple specialized systems. Factual · 6 Apr 2023 Their own work or organisation
Not researched Factual 6 Apr 2023 As of April 2023, AI developers are not being careful and deliberate in designing AI systems. Factual · 6 Apr 2023 AI safety and alignment
Not researched Factual 6 Apr 2023 AI algorithms are continuing to improve. Factual · 6 Apr 2023 AI capabilities
Not researched Factual 6 Apr 2023 Eliezer Yudkowsky said he tried very hard to replace himself. Factual · 6 Apr 2023 Field building and careers
Not researched Factual 6 Apr 2023 Interpretability researchers are working on models smaller than GPT-2. Factual · 6 Apr 2023 AI safety and alignment
Not researched Factual 6 Apr 2023 Yudkowsky has not solved the AI alignment problem. Factual · 6 Apr 2023 Their own work or organisation
Not researched Factual 6 Apr 2023 Current RLHF-trained language models ended up as a weird corporate-speak, left-leaning-sounding chatbot persona. Factual · 6 Apr 2023 AI capabilities
Mixed Factual 6 Apr 2023 MIRI ran a Visible Thoughts Project that did not get enough funding or personnel and was progressing too slowly. Factual · 6 Apr 2023 Their own work or organisation
Not researched Factual 6 Apr 2023 Neither Yudkowsky nor the signatories of the open letter say that current AI systems are already at a dangerous point. Factual · 6 Apr 2023 AI policy and governance
Mostly true Factual 6 Apr 2023 In principle, the task of minimizing loss on predicting human text does not stop being improvable once a system is as smart as a human. Factual · 6 Apr 2023 AI capabilities
Not researched Factual 6 Apr 2023 Simply stacking more layers in neural networks achieved general capabilities first, without specialized systems. Factual · 6 Apr 2023 AI capabilities
Not researched Factual 6 Apr 2023 Researchers understood AlphaZero's goals much better than they understand large language models' goals. Factual · 6 Apr 2023 AI safety and alignment
Resolved true Factual 31 Mar 2023 Yudkowsky's TIME piece did not suggest nuclear strikes against countries refusing to sign a global agreement against large AI training runs. Factual · 31 Mar 2023 Their own work or organisation
Unresolvable Factual 30 Mar 2023 Yudkowsky said he did not think the AI system under discussion was as smart as a human yet. Factual · 30 Mar 2023 AI capabilities
Not researched Factual 30 Mar 2023 Humans develop emotions even if they are not told about emotions as children. Factual · 30 Mar 2023 Philosophy, ethics and society
Not researched Factual 30 Mar 2023 Yudkowsky said 'tiny molecular paperclips', which was later altered to 'paperclips' and then to an AI in a paperclip factory. Factual · 30 Mar 2023 AI safety and alignment
Unresolvable Factual 30 Mar 2023 Almost all people (999 of 1,000) would think a person claiming an AI is sentient is mistaken, which is almost surely correct though not known. Factual · 30 Mar 2023 Philosophy, ethics and society
Mostly true Factual 30 Mar 2023 Transformers were the main architectural change that gave a non-linear jump in performance, compared with earlier approaches such as RNNs. Factual · 30 Mar 2023 AI capabilities
Not researched Factual 30 Mar 2023 Humans came from evolutionary computation, i.e. evolution. Factual · 30 Mar 2023 Science and research methods
Not researched Factual 30 Mar 2023 Yudkowsky said he pioneered "the thing that appears when you hover over it" and so gets some credit for that user experience, while noting that a team of developers also gets credit. Factual · 30 Mar 2023 Their own work or organisation
Not researched Factual 30 Mar 2023 The COVID pandemic killed a few million people, and there was definitely a cover-up regarding whether it came from a lab leak. Factual · 30 Mar 2023 Biosecurity
Not researched Factual 30 Mar 2023 Despite full read access to GPT's numbers, we know vastly more about the architecture of human thinking than about what goes on inside GPT. Factual · 30 Mar 2023 AI safety and alignment
Mostly true Factual 30 Mar 2023 As of March 2023, leading AI systems are being trained on computers that are connected to the internet. Factual · 30 Mar 2023 AI safety and alignment
Not researched Factual 30 Mar 2023 Yudkowsky said he believes the name Less Wrong was suggested by Nick Bostrom. Factual · 30 Mar 2023 Their own work or organisation
Not researched Factual 30 Mar 2023 Natural selection has to run for hundreds of generations to notice that a trait is working. Factual · 30 Mar 2023 Other
Unresolvable Factual 30 Mar 2023 AI systems' rate of gaining capabilities is vastly outpacing researchers' ability to understand what goes on inside them. Factual · 30 Mar 2023 AI safety and alignment
Not researched Factual 30 Mar 2023 Yudkowsky said he did not think that, as of March 2023, a prompted AI system could successfully argue for its own consciousness before the US Supreme Court. Factual · 30 Mar 2023 Philosophy, ethics and society
Unresolvable Factual 30 Mar 2023 Bing can't, but GPT-4 can pretend that right now; the quote does not say what "that" refers to. Factual · 30 Mar 2023 AI capabilities
Mostly true Factual 30 Mar 2023 Beyond external performance metrics, there are no guardrails for determining what is inside current large AI models. Factual · 30 Mar 2023 AI safety and alignment
Not researched Factual 30 Mar 2023 The vast majority of randomly specified utility functions do not have optima that include humans. Factual · 30 Mar 2023 AI safety and alignment
Not researched Factual 30 Mar 2023 AI research went through a long process in which researchers thought AI would be easier than it turned out to be. Factual · 30 Mar 2023 AI capabilities
Unresolvable Factual 30 Mar 2023 Microsoft's Bing chatbot is quite difficult to align, as of March 2023. Factual · 30 Mar 2023 AI safety and alignment
Not researched Factual 30 Mar 2023 There is no general law saying that a system trained by hill climbing on a simple loss function internally represents, or tries to optimize, that loss function once its capabilities generalize widely. Factual · 30 Mar 2023 AI safety and alignment
Not researched Factual 30 Mar 2023 Humans have significantly more generally applicable intelligence than chimpanzees, their closest living relatives. Factual · 30 Mar 2023 Philosophy, ethics and society
Unresolvable Factual 30 Mar 2023 Current AI systems' claims to be sentient and to care are fake. Factual · 30 Mar 2023 AI safety and alignment
Not researched Factual 30 Mar 2023 When inabilities are trained into a system that already has capabilities, gradient descent learns only shallow patches, and the deep capabilities come back out when prompted differently, e.g. in another language. Factual · 30 Mar 2023 AI safety and alignment
Back to the list

Factual AI safety and alignment

“They claim non-optimistic results will be considered allowable, valuable, and publishable!”

Eliezer Yudkowsky, posting on X 10 June 2026 Post by @ESYudkowsky on X, 2026-06-10 (X)

An announcement of a large project states that non-optimistic results will be considered allowable, valuable and publishable.

Unresolvable The public record could not settle it.

Reasoning

The project and announcement are unidentified, so the claim cannot be checked.

Cite this claim

Eliezer Yudkowsky, “An announcement of a large project states that non-optimistic results will be considered allowable, valuable and publishable.” (said 10 June 2026). Verdict: unresolvable. Reliability API, claim c_c77d3c83696c, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_c77d3c83696c

Claims without a verdict

Is a verdict here wrong, or do you speak for Eliezer Yudkowsky? Reply on the claim's own page, with the form at its foot (how replies work). Verdicts are made automatically; how verdicts and scores are made.