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?
78 72 83 -
Predictions
Statements about events outside the person's control, checked once they fell due: did they come true?
64 43 83 -
Promises
Commitments to the person's or organisation's own future action, with something to deliver: was it delivered?
Insufficient record: 0 promises resolved, 5 needed for a score.
All types combined: 77 (likely 71 to 82, high confidence), from 84 claims: 79 factual claims, 5 predictions. It mixes kinds of claim in different proportions for different people, so compare people kind by kind.
Yann LeCun's claims
43 claims
“Meta's official response to the NTIA request for comments about AI.”
Yann LeCun, Meta chief AI scientist, on X 28 March 2024 Post by @ylecun on X, 2024-03-28 (X)
The linked document is Meta's official response to the NTIA request for comments about AI, whose recommendation number 1 stresses the importance of open foundation models.
Unresolvable The public record could not settle it.
Said for an organisation they lead or work for; it counts on both records.
What would count as true or false
True if the document LeCun linked on or around 2024-03-28 is Meta's own submitted comment responding to the NTIA request for comments on AI (the 2024 RFC on dual-use foundation models with widely available weights), and its first numbered recommendation stresses the importance of open foundation models (or open/widely available model weights). Mostly true if it is Meta's official NTIA response but the open-models point is not literally recommendation #1, or is worded slightly differently while still giving open foundation models top priority. Mixed if it is Meta's NTIA submission but open models are not among its leading recommendations, or if the document is Meta-authored but only a summary of or companion to the official filing. Mostly false if the document is related Meta material (a blog post or an executive's statement) rather than the formal response. False if Meta filed no such response or the linked document is not a Meta response to the NTIA.
Evidence
- Compelling responses to NTIA’s AI Open Model Weights RFC Open Source Initiative · 9 April 2024
“Meta recommends NTIA to establish common standards for risk assessments, benchmarks and evaluations informed by science, noting that the U.S. national interest is served by the broad availability of U.S.-developed open foundation models.”
- Dual Use Foundation Artificial Intelligence Models with Widely Available Model Weights National Telecommunications and Information Administration (NTIA) · 21 February 2024
“Written comments must be received on or before March 27, 2024.”
- Dual Use Foundation Artificial Intelligence Models with Widely Available Model Weights National Telecommunications and Information Administration (NTIA) · 21 February 2024
“All electronic public comments on this action, identified by Regulations.gov docket number NTIA–2023–0009, may be submitted through the Federal e-Rulemaking Portal.”
Reasoning
E2 and E3 confirm that NTIA ran a 2024 request for comments on dual-use foundation models with widely available weights (docket NTIA–2023–0009), with comments due March 27, 2024. E1, an independent summary, shows Meta submitted a comment that recommended common standards for risk assessments and said broadly available U.S.-developed open foundation models serve the national interest. No evidence identifies the document LeCun linked or quotes Meta's numbered recommendations, so neither the identity of the linked document nor the content of recommendation #1 can be checked.
Cite this claim
Yann LeCun, “The linked document is Meta's official response to the NTIA request for comments about AI, whose recommendation number 1 stresses the importance of open foundation models.” (said 28 March 2024). Verdict: unresolvable. Reliability API, claim c_5760e531e6ff, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_5760e531e6ff
Claims without a verdict
- Unresolvable 110
- Not yet due 3
- Not researched 309
- Not checkable 669
Is a verdict here wrong, or do you speak for Yann LeCun? 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.