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?
74 67 80 -
Predictions
Statements about events outside the person's control, checked once they fell due: did they come true?
62 43 81 -
Promises
Commitments to the person's or organisation's own future action, with something to deliver: was it delivered?
Insufficient record: 1 promise resolved, 5 needed for a score.
All types combined: 73 (likely 66 to 79, high confidence), from 72 claims: 66 factual claims, 6 predictions. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.
Michael Burry's claims
30 claims
“I have never said that LLMs would stop improving”
Michael Burry, investor, in Substack chat 27 October 2026 Burry says Anthropic's IPO filing changes nothing about AI bear case (Stocktwits)
Michael Burry has never said that LLMs would stop improving.
Mostly true True in substance, wrong 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
Test: whether any public statement by Burry made before 27 October 2026 (X/Twitter posts, Substack, interviews, letters) said that large language models would stop improving or hit a ceiling or plateau. True: a review of his public statements finds no such statement, and his comments on LLMs concern other issues (valuation, capex, depreciation, diminishing returns). Mostly true: no explicit statement, but a remark loosely suggesting a plateau that he qualified at the time. Mixed: ambiguous statements that could reasonably be read either way. Mostly false: he said LLM progress would stall or plateau in substance without using those exact words. False: a verifiable statement in which he explicitly said or predicted that LLMs would stop improving.
Evidence
- Michael Burry Says Anthropic’s IPO Filing Changes Nothing About His AI Bear Case — Flags Financial Risks Beyond The Tech Asianet News Newsable · 27 October 2026
“Everyone since 2024/5 said LLMs will hit a wall/plateau but it really didn’t and is still exponentially improving.”
- ‘Big Short’ investor Michael Burry says AI warnings are self-serving Washington Examiner · 14 September 2026
“1. LLMs are not AI and won't be AGI. There is nothing AI to slow down. 2. Competition is coming up fast, slowing benefits incumbents.”
- Michael Burry (@michaeljburry) Substack · 14 September 2026 Own record
“1. LLMs are not AI and won't be AGI. There is nothing AI to slow down.”
- Michael Burry (@michaeljburry) Substack · 14 September 2026 Own record
“I propose it is hard to slow down AI when what is advancing is increasingly sophisticated language models at the frontier.”
- Michael Burry Warns AI Models May Never Reach AGI As Synthetic Data Risks Triggering Model Collapse: 'Understanding Cannot Exist Unless...' Benzinga / Yahoo Finance · 28 September 2026
“Burry also made a broader argument about artificial general intelligence, writing that "LLMs cannot attain understanding (AGI)" because "understanding cannot exist unless reason first exists without language."”
- Michael Burry Warns AI Models May Never Reach AGI As Synthetic Data Risks Triggering Model Collapse: 'Understanding Cannot Exist Unless...' Benzinga / Yahoo Finance · 28 September 2026
“This is to my point about compression being inevitable as human knowledge is too small for what we are building.”
Reasoning
None of the Burry statements in the evidence explicitly says that LLMs would stop improving. In September 2026 he argued that LLMs are not AI and will not become AGI or attain understanding (E4, E5, E7), which is a claim about what they can ever do, not about progress stopping. In the same note he described frontier language models as advancing and becoming more sophisticated (E6). His point that "compression" is "inevitable as human knowledge is too small for what we are building" (E8) loosely suggests a limit, but it does not state that LLM progress would stop. The evidence covers only a sample of his public statements, not a full review.
Cite this claim
Michael Burry, “Michael Burry has never said that LLMs would stop improving.” (said 27 October 2026). Verdict: mostly true. Reliability API, claim c_ed0d9e795b8d, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_ed0d9e795b8d
Claims without a verdict
Is a verdict here wrong, or do you speak for Michael Burry? 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.