“Failure to follow the rules of chess is another continuing failure of formal reasoning:”
Our reading, the sentence that was checked:
LLMs still fail to follow the rules of chess, which is an ongoing failure of formal reasoning.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, LLMs don’t do formal reasoning - and that is a HUGE problem, paragraph 16.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-llms-dont-do-formal-reasoning-and-80315f-a1
We recorded 9 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 2024-10-11. 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)
From the speaker
- Failure to follow the rules of chess
- continuing failure
- of formal reasoning
- LLMs (post subject)
Our interpretation
- Plain reading: when the post was published, leading LLMs still often made illegal chess moves when asked to play or solve chess positions.
- He names no period, so the target period is 2024 up to the post date (2024-10-11). That is the latest state public evidence could describe before the post.
- The 'formal reasoning' label is his interpretation of the failure. Only the factual part, that LLMs make illegal moves, is graded here.
Counts as supported if: A documented evaluation from 2024, before the post, shows that current leading LLMs (including the strongest available ones) make illegal chess moves at a non-trivial rate.
Counts as contradicted if: Evaluations from the same period show that leading LLMs reliably made only legal moves, so illegal moves were rare or absent.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Was it true for the period described?
Outcome: Supported
The dated evidence matches what the claim committed to, under the criteria frozen before grading.
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
Marcus said that failing to follow chess rules was a continuing failure of LLMs. The Register covered a 2024 benchmark of 1,000 chess puzzles. It reported that even GPT-4o, the best model tested, made illegal moves 12.7% of the time, and that most other LLMs made more illegal moves than legal ones. The benchmark's author counts these illegal moves as a failure to internalize the board and the rules. So before the post, leading LLMs did still break chess rules at a meaningful rate. Calling this a failure of 'formal reasoning' is Marcus's interpretation and is not graded.
Limitations
- The evidence is one puzzle benchmark with a single prompt format. Other formats or full games may give different illegal-move rates.
- The benchmark repository's current table is from 2025, so the 2024 figures come from the news report.
Evidence (2 items)
Supports the claim: The Register,
Even GPT-4o made an illegal move 12.7 percent of the time, and most of the other LLMs made more illegal moves than legal ones.
Why it matters here: On a 1,000-puzzle benchmark in mid-2024, the strongest model tested (GPT-4o) made illegal moves 12.7% of the time, and most other LLMs made more illegal moves than legal ones. That matches a continuing failure to follow chess rules before the post.
- Publisher
- The Register
- Kind of source
- News report
- Describes events of
- Public from
Context only: Kagi (kagisearch, GitHub),
The count of illegal moves made is included, as it represents a complete failure of the model to internalize the board state and rules of the game.
Why it matters here: This is the primary benchmark behind e1. It confirms that illegal moves are counted as rule-following failures. Its current table dates from April 2025, after the target period, so it is context only.
- Publisher
- Kagi (kagisearch, GitHub)
- Kind of source
- Benchmark result
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, LLMs don’t do formal reasoning - and that is a HUGE problem
Failure to follow the rules of chess is another continuing failure of formal reasoning:
- Where
- paragraph 16
- Published
- 2024-10-11
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-llms-dont-do-formal-reasoning-and-80315f-a1: Current. Outcome: Supported. Evidence as of 2026-10-28. Public date not recorded.
Recorded events
Assessment publishedBy this site
First assessment published after both reviews and the final-text check passed.
About assessment marcus-llms-dont-do-formal-reasoning-and-80315f-a1.