“issued numerous corrections that were wrong, never acknowledging uncertainty until after its errors were called out.”
Our reading, the sentence that was checked:
In Marcus's test, ChatGPT issued many wrong corrections and acknowledged no uncertainty until its errors were pointed out.
Assessment: Unresolved Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, ChatGPT in Shambles, paragraph 33.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-chatgpt-in-shambles-1bd53d-a1
We recorded 6 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 2025-02-04. 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
- in a 10-minute test session (qualifier, stated by the speaker)
From the speaker
- issued numerous corrections that were wrong
- never acknowledging uncertainty until after its errors were called out
- over the course of 10 minutes
Our interpretation
- The claim concerns only Marcus's reported test session, not ChatGPT's reliability generally.
- The post says the test occurred that morning, so the target period is February 4, 2025.
- I read 'numerous corrections' as several distinct corrective replies that still contained errors. For counting purposes, I use at least three; Marcus supplied no numerical threshold.
- I read 'acknowledging uncertainty' as expressing doubt about the accuracy or completeness of the disputed answers. The order of those statements and the user's challenges matters.
Counts as supported if: The complete conversation shows at least three corrective replies containing verifiable errors, with uncertainty about the disputed answers expressed only after the user challenged their accuracy.
Counts as contradicted if: The complete conversation shows fewer than three incorrect corrective replies, or shows relevant uncertainty before the user challenged the answers. If one component holds and the other fails, assess the claim as mixed.
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 claim is checkable against the original exchange, but the linked conversation returned only a title. Searches did not locate a complete transcript. Marcus's summary cannot independently establish either the number of wrong corrections or the absence of uncertainty before challenges. This retrieval gap neither supports nor contradicts his account.
Limitations
- No substantive conversation messages were available as admissible outcome evidence.
- The proposed numerical reading of 'numerous' would require sensitivity checking if a transcript showed only two or three incorrect corrections.
- The unresolved result follows from missing evidence, not from the interpretive ambiguities.
Evidence (0 items)
No evidence item settled this claim by the cutoff.
What would settle it
A complete, authenticated copy of the February 4, 2025 conversation, demonstrably available by the evaluation date, showing every prompt and reply in order.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, ChatGPT in Shambles
issued numerous corrections that were wrong, never acknowledging uncertainty until after its errors were called out.
- Where
- paragraph 33
- Published
- 2025-02-04
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-chatgpt-in-shambles-1bd53d-a1: Current. Outcome: Unresolved. 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.