“There are some applications where the current systems are reliable enough to be practically useful, though not nearly as many as is often claimed.”
Our reading, the sentence that was checked:
Current systems are reliable enough to be useful in some applications, but far fewer than often claimed.
Assessment: Unresolved Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Reports of the birth of AGI are greatly exaggerated, paragraph 23.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-reports-of-the-birth-of-agi-are-greatly-2108c3-a1
We recorded 11 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 2023-10-18. 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
- some applications
- current systems
- reliable enough to be practically useful
- not nearly as many as is often claimed
Our interpretation
- Current systems means the next-token-prediction systems discussed in the surrounding paragraph.
- I assess October 1–18, 2023, the current calendar month through publication.
- The ordinary reading combines an existence claim about useful applications with a substantial overstatement claim about common contemporary descriptions of their usefulness.
- Testing the second part requires an identifiable, reasonably representative collection of contemporary application claims. I do not invent a count or treat every proposed application as a claim of proven reliability.
Counts as supported if: Evidence establishes useful applications and shows that a representative set of contemporary claims substantially exceeds the applications reliably usable in practice.
Counts as contradicted if: A representative comparison shows that contemporary claims generally matched demonstrated practical reliability, or evidence disproves the existence of any practically useful applications.
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 research provides concrete examples of usefulness and limitations. It does not establish the much stronger comparison that reliably useful applications were far fewer than commonly claimed. The speaker identifies neither the claims nor the applications to count. Qualitative demonstrations and proposed applications also cannot automatically be treated as promises of production reliability. The full proposition therefore remains unresolved.
Limitations
- The existence of useful assistance is easier to test than the claimed extent of overstatement.
- The cited contextual records precede the selected October observation window.
- A handful of successful or unsuccessful tasks cannot establish a representative application count.
Evidence (2 items)
Context only: Boston Consulting Group,
nearly all participants improved
Why it matters here: The experiment documents useful assistance alongside worse results on another task. It predates the target window and supplies neither an application count nor a representative assessment of public claims.
- Publisher
- Boston Consulting Group
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Context only: arXiv; Zhengyuan Yang and colleagues,
emerging application scenarios
Why it matters here: This contemporary report explores broad capabilities and possible applications through qualitative examples. Its discussion of emerging scenarios is not equivalent to claiming that every scenario was already reliably deployable.
- Publisher
- arXiv; Zhengyuan Yang and colleagues
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
What would settle it
A documented sample of prevalent claims available by October 18, 2023, paired with evaluations of the same applications, models and supervision requirements, showing whether usefulness was substantially overstated.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Reports of the birth of AGI are greatly exaggerated
There are some applications where the current systems are reliable enough to be practically useful, though not nearly as many as is often claimed.
- Where
- paragraph 23
- Published
- 2023-10-18
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-reports-of-the-birth-of-agi-are-greatly-2108c3-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-reports-of-the-birth-of-agi-are-greatly-2108c3-a1.