“It’s way too much charity to deep learning to claim that there is any kind of direct link.”
Our reading, the sentence that was checked:
The evidence does not support claiming any direct link between deep learning models' inner workings and the human brain.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Three baffling claims about AI and machine learning in four days, statistical errors in top journals, and claims from Yann LeCun that you should not believe., paragraph 81.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-too-much-benefit-of-the-doubt-7a87fd-a1
We recorded 13 claims from this post. See every claim from this post.
The wording above is the version this site observed on 2026-10-29. The source may have been edited after 2022-10-16. 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
- way too much charity to deep learning
- any kind of direct link
- actually functioning in the same way in both
Our interpretation
- Assess the cited study and its author's public explanation, from publication on 2022-09-29 through the 2022-10-11 thread.
- In the surrounding argument, read 'direct link' as demonstrated shared internal mechanisms, rather than any measurable statistical association.
- This reading does not imply that the study supplies no evidence relevant to shared representations.
Counts as supported if: The reported link consists of predictive associations and model comparisons, without establishing that the brain and GPT-2 implement the same internal mechanisms.
Counts as contradicted if: The study establishes the shared internal mechanism at issue through evidence that distinguishes it from alternative explanations of the observed associations.
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
The evidence supports Marcus's narrower point that a shared internal mechanism was not established. The author's thread also reports meaningful correspondences and model manipulations, so it does support an empirical link. Calling that a 'direct link' need not assert identical mechanisms. The outcome therefore depends on the contextual mechanistic reading.
Limitations
- This does not establish that shared mechanisms are absent.
- A literal denial of any empirical link would not be supported.
- The cited thread is the researcher's own summary, reproduced by a third party.
Evidence (1 item)
Supports the claim: Charlotte Caucheteux, reproduced by Thread Reader,
And it works:
Why it matters here: The author's account describes mapping, correlations and effects of changing model depth and attention. Inferring that these do not establish identical biological mechanisms is the evaluator's assessment, not an explicit concession in this quotation.
- Publisher
- Charlotte Caucheteux, reproduced by Thread Reader
- Kind of source
- Secondary summary
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Three baffling claims about AI and machine learning in four days, statistical errors in top journals, and claims from Yann LeCun that you should not believe.
It’s way too much charity to deep learning to claim that there is any kind of direct link.
- Where
- paragraph 81
- Published
- 2022-10-16
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-too-much-benefit-of-the-doubt-7a87fd-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-too-much-benefit-of-the-doubt-7a87fd-a1.