Pundit ClaimsAI claims by Gary Marcus and Ed Zitron, with the evidence

“So Nestler’s experiment does not contradict the finding of the report; it corroborates it. o3, yet again, produced an invalid answer to this problem.”

Reports of LLMs mastering math have been greatly exaggerated, paragraph 9. The full statement is at the source.

Our reading, the sentence that was checked:

The o3 proof of USAMO 2025 problem 2 posted by Nestler was invalid, corroborating the report.

Assessment: Unresolved Evidence as of

Where and when Gary Marcus wrote it

Marcus on AI, Reports of LLMs mastering math have been greatly exaggerated, paragraph 9.

Said
Source observed
live page, full text, public
Evidence as of
the cutoff for this assessment
Assessment recorded
assessment marcus-reports-of-llms-mastering-math-have-05b28b-a1

We recorded 11 claims from this post. See every claim from this post.

The wording above is the version this site observed on 2026-10-28. The source may have been edited after 2025-04-05. This page does not claim the wording was captured then.

What the claim commits to

Kind
Observation
Deadline, as the speaker wrote it
None stated. This site does not add one.
Deadline state when assessed
No horizon
Deadline state on 2026-10-29
No horizon (the release’s reference date)

Conditions

  • USAMO 2025 problem 2 (qualifier, stated by the speaker)

From the speaker

  • Nestler’s experiment
  • o3, yet again, produced an invalid answer
  • this problem
  • it corroborates it

Our interpretation

  • The object being assessed is the particular proof of USAMO 2025 problem 2 that Marcus says Nestler posted, not other outputs on that problem.
  • Invalid means a substantive false step, missing argument or unjustified inference that prevents the posted proof from establishing the result.
  • Corroboration means another instance of presenting an invalid proof as a solution, not an independent replication of the report's aggregate scores.
  • The target period runs from the contest's March 19 start, identified in the post, through the April 5 publication. The exact date of Nestler's output is not supplied.

Counts as supported if: Recover the identifiable posted proof and establish a substantive mathematical defect through a checkable analysis or independent expert review.

Counts as contradicted if: Recover that same proof and establish that it is complete and valid, including the step alleged to be defective.

Retrospective: these criteria were written after the outcome was public.

Current assessment

Was it true for the period described?

Outcome: Unresolved

The record was checked, and the evidence available by the cutoff does not settle it.

Scored on a plain reading. The speaker named no measure for this claim, so it was judged on the reading an informed reader would take, stated below under our interpretation. Both reviews had to find that reading fair.

Evidence as of
Deadline state when assessed
No horizon
Assessment recorded
Policy version
2026-10-28.7

Rationale

The searches did not recover Nestler's specific proof or an independent analysis of it. The USAMO study documents failures by other evaluated outputs, but that cannot settle this allegation. Marcus's own statement is insufficient evidence.

Limitations

  • Search and page-access limitations may hide a public original or review.
  • Failure to recover the proof is not evidence that Marcus's criticism was wrong.

Evidence (0 items)

No evidence item settled this claim by the cutoff.

What would settle it

An authenticated copy of Nestler's complete original output, followed by a checkable mathematical review identifying a substantive defect or confirming the proof's validity.

Where it was said (1)

Said once in the sources this register covers.

  • : Marcus on AI, Reports of LLMs mastering math have been greatly exaggerated

    So Nestler’s experiment does not contradict the finding of the report; it corroborates it. o3, yet again, produced an invalid answer to this problem.

    Published
    2025-04-05
    Source observed
    , live page
    Access
    full text, public

History

Assessment versions

Recorded events

  1. Assessment publishedBy this site

    First assessment published after both reviews and the final-text check passed.

    About assessment marcus-reports-of-llms-mastering-math-have-05b28b-a1.