“And all this from a company thought (or claimed) that GPT-5 was going to be tantamount to AGI (spoiler alert: it wasn’t)”
Our reading, the sentence that was checked:
GPT-5 was not tantamount to AGI, although OpenAI thought or claimed that it would be.
Assessment: Unresolved Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Peak absurdity, Part II, paragraph 6.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-peak-absurdity-part-ii-60174f-a1
We recorded 3 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 2026-04-15. 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
- OpenAI thought (or claimed) (qualifier, stated by the speaker)
From the speaker
- a company thought (or claimed)
- GPT-5 was going to be tantamount to AGI
- it wasn’t
Our interpretation
- This contains two assertions: an earlier OpenAI expectation or claim, and GPT-5's failure to reach that level by the September 2025 quotation.
- The ordinary capability reading is broad, human-level general intellectual performance, rather than success on an isolated benchmark.
- Company expectations require an attributable statement or reliable evidence of internal expectations. Outside speculation alone is insufficient.
- The historical interval has no specified beginning because the alleged earlier expectation is undated.
- The phrase 'thought (or claimed)' is part of the historical assertion, not a conditional forecast antecedent.
Counts as supported if: Evidence establishes both that OpenAI expected or claimed GPT-5 would amount to AGI and that the released GPT-5 fell short under an ordinary, stated understanding of AGI.
Counts as contradicted if: Reliable evidence refutes the attributed expectation or claim, or establishes that GPT-5 had reached the relevant general capability level. Failure to find an earlier statement alone is not refutation.
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 launch admission supports the capability clause. The earlier attribution remains unverified: the located story reports an outsider's account of unnamed employees and explicitly limits its scope. That does not establish the whole compound claim. It also does not refute it, so a mixed verdict would overstate the evidence.
Limitations
- Support for one clause is insufficient to support the whole claim.
- Corporate statements about AGI are relevant evidence, not an independent universal capability test.
Evidence (3 items)
Supports the claim: The Guardian,
not reached
Why it matters here: Supports capability clause.
- Publisher
- The Guardian
- Kind of source
- News report
- Describes events of
- Public from
Context only: OpenAI,
internal benchmark
Why it matters here: OpenAI reports expert-comparable or better reasoning performance in roughly half the evaluated cases across more than 40 occupations. This describes substantial capabilities, but does not establish AGI or an earlier promise to deliver it.
- Publisher
- OpenAI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Context only: Gizmochina,
not publicly discussed
Why it matters here: The article attributes the expectation to Siqi Chen's account of some employees, and reports his clarification that it was not a consensus. This is a possible origin of Marcus's attribution, but does not verify a corporate position or connect those expectations securely to the eventual release.
- Publisher
- Gizmochina
- Kind of source
- News report
- Describes events of
- Public from
What would settle it
An attributable pre-release OpenAI statement or authenticated internal record connecting GPT-5 to the claimed AGI expectation, assessed alongside contemporaneous capability evidence under a stated ordinary definition.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Peak absurdity, Part II
And all this from a company thought (or claimed) that GPT-5 was going to be tantamount to AGI (spoiler alert: it wasn’t)
- Where
- paragraph 6
- Published
- 2026-04-15
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-peak-absurdity-part-ii-60174f-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.