“We are going to see a lot of this – misuses of ChatGPT in which people trust it, and even trumpet it—when it’s flat-out wrong.”
Our reading, the sentence that was checked:
There will be many cases of people trusting and promoting ChatGPT outputs that are flat-out wrong.
Assessment: Unresolved No deadline stated on 2026-10-29 Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Scientists, please don’t let your chatbots grow up to be co-authors, paragraph 9.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-scientists-please-dont-let-your-chatbots-ec90d4-a1
We recorded 15 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 2023-01-14. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Forecast
- Deadline, as the speaker wrote it
- None stated. This site does not add one.
- Deadline state when assessed
- No deadline stated
- Deadline state on 2026-10-29
- No deadline stated (the release’s reference date)
From the speaker
- We are going to see a lot of this
- misuses of ChatGPT
- people trust it, and even trumpet it
- when it’s flat-out wrong
Our interpretation
- The ordinary reading predicts a recurring public pattern of people relying on demonstrably false ChatGPT outputs and presenting them to others as credible.
- Public submission of purported authorities counts as presenting the output as credible; explicit praise of ChatGPT is a stronger example, not a requirement in every episode.
- “A lot” requires evidence beyond isolated anecdotes, but Marcus supplies no numerical threshold.
- Only incidents attributable to ChatGPT count. Undifferentiated generative-AI totals cannot settle the claim.
Counts as supported if: A documented collection establishes many independent post-publication episodes of reliance on and public presentation of demonstrably false ChatGPT output.
Counts as contradicted if: The alleged qualifying episodes are disproved or fail attribution and reliance checks. Mere absence of examples by an evaluator-selected date cannot contradict this forecast without a deadline.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Did the forecast come true?
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 deadline stated
- Assessment recorded
- Policy version
- 2026-10-28.7
Rationale
The records establish a clear qualifying example and provide substantial indications of recurrence. But the larger count measures matters naming ChatGPT, and its underlying database warns that naming a tool does not establish responsibility for the false material. The records verified here therefore do not meet the stated early-resolution rule. With no deadline, this is not a failed forecast.
Limitations
- This is a finding about the evidence verified here, not a finding that numerous qualifying incidents did not occur.
- The volume requirement and meaning of public promotion remain qualitative.
Evidence (3 items)
Supports the claim: High Court of Justice of England and Wales,
ChatGPT
Why it matters here: The judgment recounts a lawyer presenting fabricated ChatGPT authorities and returning to the tool when challenged. This establishes a qualifying episode, but one episode cannot settle “a lot”.
- Publisher
- High Court of Justice of England and Wales
- Kind of source
- Regulatory filing
- Describes events of
- Public from
Context only: AI Vortex,
ChatGPT 118
Why it matters here: The dated snapshot indicates many court matters naming ChatGPT. Its measure is tool disclosure, however, rather than verified attribution of each false output and the claimed human reliance.
- Publisher
- AI Vortex
- Kind of source
- Secondary summary
- Describes events of
- Public from
Context only: Damien Charlotin,
specific tool
Why it matters here: The underlying database expressly cautions against treating tool names as responsibility findings. This prevents automatically converting the aggregate into 118 qualifying ChatGPT incidents.
- Publisher
- Damien Charlotin
- Kind of source
- Independent measurement
- Describes events of
- Public from
What would settle it
A dated, reviewable collection of numerous independent incidents, with underlying records establishing that people relied on and publicly presented demonstrably false ChatGPT output.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Scientists, please don’t let your chatbots grow up to be co-authors
We are going to see a lot of this – misuses of ChatGPT in which people trust it, and even trumpet it—when it’s flat-out wrong.
- Where
- paragraph 9
- Published
- 2023-01-14
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-scientists-please-dont-let-your-chatbots-ec90d4-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.
About assessment marcus-scientists-please-dont-let-your-chatbots-ec90d4-a1.