““Work well” remains a tendentious claim; they still cannot be trusted, make random mistakes, have no basis in factuality.”
Our reading:
The AI systems Hinton says 'work well' still cannot be trusted, make random mistakes, and have no basis in factuality, so calling them 'working well' is tendentious.
Assessment: Can’t be checked
What happened
Not graded on evidence
Outcome: Can’t be checked
An opinion, a recommendation, or too vague to test. Listed, not scored.
Evidence up to . Checked .
Why
The reliability warning has empirical backing, but the combined wording does not identify one testable proposition. Factuality studies can find both accurate information and serious errors in the same system. Whether that supports or contradicts 'no basis in factuality' depends on an interpretation Marcus did not clearly supply. Calling this mixed would conceal that ambiguity rather than resolve it.
Caveats
- This does not reject the narrower claim that contemporary language models made factual errors.
- The cited study is context, not decisive evidence for the combined wording.
Evidence (1 item)
Claims like this aren’t scored, so there is no evidence to show.
How we judged it
- Kind of claim
- Observation
- Deadline
- None given. We don’t add one.
Conditions
- the systems Hinton described as working well (their words)
Where we had to interpret
- The context moves between neural networks generally and generative AI. It does not identify one consistently defined system population.
- 'No basis in factuality' could mean an absence of factual knowledge or an absence of a dependable guarantee of truth. Those are materially different claims.
- 'Cannot be trusted' lacks a specified use or required level of accuracy, while 'work well' is expressly evaluative.
- No single categorical contract can resolve these differences without adding a meaning the speaker did not clearly supply.
We don’t grade claims like this. We only check the quote and who said it.
What would settle it
Specify the systems and uses, and clarify whether factual basis means factual knowledge, external grounding, or dependable factual correctness.
We found 22 claims in this post. See them all.
Full record
Source and dates
Marcus on AI, What was 60 Minutes thinking, in that interview with Geoff Hinton?, paragraph 40.
- Said
- We read the source
- live page, full text, public
- Evidence up to
- Checked
We read this wording on 2026-10-29. The post may have been edited after 2023-10-10.
Where it was said (1)
Said once in the posts we read.
: Marcus on AI, What was 60 Minutes thinking, in that interview with Geoff Hinton?
“Work well” remains a tendentious claim; they still cannot be trusted, make random mistakes, have no basis in factuality.
- Where
- paragraph 40
- Published
- 2023-10-10
- We read it
- , live page
- Access
- full text, public
History
Versions
- Current: Can’t be checked. Evidence up to 2026-10-28.
What happened to this record
PublishedBy us
Published after both AI reviews and the final check passed.
Held backBy us
A reviewer said our test wasn’t fair to what was actually said.