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

8 claims, sorted by newest said. Choose a claim to read it in full.
VerdictKindSaidClaimTopic
Mixed Prediction 5 Mar 2024 If a rival AI developer beats GPT-4 on general intelligence, OpenAI will be forced to release GPT-4.5. Prediction · 5 Mar 2024 AI capabilities
Mostly false Prediction 6 Apr 2023 Paul Christiano will probably win his forecast disagreement with Yudkowsky on an AI getting IMO gold by 2025, with prediction-market odds then running around 30%. Prediction · 6 Apr 2023 AI timelines and forecasting
Resolved false Prediction 6 Apr 2023 The current AI developers will not try the alignment approach he described (filtering training data to nice examples and training on it). Prediction · 6 Apr 2023 AI safety and alignment
Resolved true Prediction 6 Apr 2023 AI alignment will not be solved within a few years. Prediction · 6 Apr 2023 AI safety and alignment
Resolved true Prediction 6 Apr 2023 GPT-5 will not end the world, with more than 50% of Yudkowsky's probability on that outcome. Prediction · 6 Apr 2023 AI safety and alignment
Mostly true Prediction Feb 2023 Yudkowsky said he felt quite confident that the upper range of the untapped potential of ChatGPT-style systems is insufficient to outsmart all living humans and carry out the catastrophe scenario he worries about. Prediction · Feb 2023 AI capabilities
Resolved true Prediction 2008 Yudkowsky doubted that a thousandfold increase in AI hardware would buy anything like a thousandfold increase in effective intelligence, while noting that such doubt is not reliable without an analytical calculation. Prediction · 2008 AI compute, hardware and data
Mostly true Prediction 2008 Most AI researchers will not want to make unFriendly AIs. Prediction · 2008 AI safety and alignment
Back to the list

Prediction AI timelines and forecasting

“So probably Paul’s going to win, but slight moral victory.”

Eliezer Yudkowsky, interviewee 6 April 2023 Eliezer Yudkowsky - Why AI Will Kill Us, Aligning LLMs, Nature of Intelligence, SciFi, & Rationality (Dwarkesh Podcast)

Paul Christiano will probably win his forecast disagreement with Yudkowsky on an AI getting IMO gold by 2025, with prediction-market odds then running around 30%.

Mostly false False in substance, right in a detail.

The second judge would not resolve this claim; the verdict shown is the first judge's.

What would count as true or false

Paul Christiano 'wins' the bet if no AI system achieves an International Mathematical Olympiad gold-medal standard by the end of 2025, under the terms of their 2021 bet. True: no AI achieved IMO gold-level performance through IMO 2025 under the bet's conditions. Mostly true: the only claimed gold-level results fall clearly outside the bet's terms (for example, extra time or human-formalized problems) and are not accepted as satisfying it. Mixed: a gold-level claim exists but whether it satisfies the bet is genuinely disputed. Mostly false: an AI met the gold standard with only minor deviations from the bet's conditions. False: an AI achieved IMO gold-medal-level performance by 2025 in a way that satisfies the bet, as acknowledged by the participants or by credible verification.

Evidence

  1. Google and OpenAI Get 2025 IMO Gold — LessWrong LessWrong (post by Zvi Mowshowitz, 'Don't Worry About the Vase') · 22 July 2025 Own record
    “So I think we have Paul at <8%, Eliezer at >16% for AI made before the IMO is able to get a gold (under time controls *etc.* of grand challenge) in one of 2022-2025.”
  2. Google and OpenAI Get 2025 IMO Gold — LessWrong LessWrong (post by Zvi Mowshowitz, 'Don't Worry About the Vase') · 22 July 2025 Own record
    “The results were not super clear cut if you look at the details, as Christiano was in large part doubting that the hardest problem would be solved and indeed the hardest problem was #6 and was not solved, but a gold medal was still achieved.”
  3. Advanced version of Gemini with Deep Think officially achieves gold-medal standard at the International Mathematical Olympiad Google DeepMind · 21 July 2025
    “The IMO have confirmed that our submitted answers are complete and correct solutions. It is important to note that their review does not extend to validating our system, processes, or underlying model”
  4. AI at IMO 2025: a round-up Xena Project (Kevin Buzzard, Imperial College London mathematician) · 3 August 2025
    “It is very exciting to see progress in the mathematical capabilities of AI models, but we would like to be clear that the IMO cannot validate the methods, including the amount of compute used or whether there was any human involvement, or whether the results can be reproduced.”

Reasoning

Google DeepMind reported gold-medal-standard performance at IMO 2025, and the IMO confirmed its submitted solutions were complete and correct, though its review did not extend to validating the system, processes or model (E3); a LessWrong post says a gold medal was achieved (E2). This is against Christiano's side of the bet, which concerned gold under grand-challenge time controls and conditions (E1). The IMO said it could not validate the methods, compute or human involvement (E3, E4), and E2 calls the outcome not fully clear-cut because problem 6 went unsolved. The evidence does not show whether the bet's conditions were fully met or whether either participant acknowledged a result, so gold was reached with some uncertainty about the bet's terms.

Cite this claim

Eliezer Yudkowsky, “Paul Christiano will probably win his forecast disagreement with Yudkowsky on an AI getting IMO gold by 2025, with prediction-market odds then running around 30%.” (said 6 April 2023). Verdict: mostly false. Reliability API, claim c_2b396ed98e67, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_2b396ed98e67

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.