“Almost exactly what I predicted we would see with GPT-4, back on Christmas Day 2022.”
Our reading, the sentence that was checked:
Altman's description of o1 as 'still flawed, still limited, seem more impressive on first use' is almost exactly what Marcus predicted we would see with GPT-4 on Christmas Day 2022.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Gary Marcus vs Sam Altman on GPT-o1 aka Strawberry, paragraph 14.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-gary-marcus-vs-sam-altman-on-gpt-de96b1-a1
We recorded 5 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 2024-09-12. 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
- almost exactly (qualifier, stated by the speaker)
From the speaker
- Almost exactly what I predicted
- with GPT-4
- Christmas Day 2022
- Still flawed, still limited, seem more impressive on first use
Our interpretation
- Compare the substance of the December 25, 2022 prediction with Altman's September 12, 2024 description.
- Read almost exactly as agreement on persistent flaws, limitations, and an initially inflated impression; identical wording is unnecessary.
- This is a claim about correspondence between two statements, not proof that all seven GPT-4 predictions came true.
- Marcus's earlier article is used only to establish the prediction being compared, not as independent evidence that his model-performance claims were correct.
Counts as supported if: The dated prediction anticipates persistent flaws and limitations becoming clearer after an impressive initial reception, and Altman's later statement makes the same substantive points.
Counts as contradicted if: The original prediction materially omits or conflicts with those points, or the attributed Altman statement is materially inaccurate.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Was it true for the period described?
Outcome: Supported
The dated evidence matches what the claim committed to, under the criteria frozen before grading.
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 comparison holds in substance. Marcus's December 2022 passage anticipated initial excitement followed by recognition of persistent problems, and his predictions specified continuing errors and limitations. Altman's September 2024 statement describes those same broad features. The wording and models differ, and collective scientific reassessment differs from personal experience, but those differences do not defeat the ordinary meaning of almost exactly. This finding does not establish that Marcus's complete GPT-4 forecast was correct.
Limitations
- The agreement is thematic, not verbatim.
- Altman's statement concerns o1; Marcus's prediction concerned GPT-4.
- The assessment establishes correspondence between statements, not independent confirmation of their claims about model performance.
Evidence (2 items)
Supports the claim: Sam Altman / X,
still flawed, still limited
Why it matters here: Altman's original statement contains all three elements of the comparison: continuing flaws, continuing limitations, and a stronger first impression than sustained use supports.
- Publisher
- Sam Altman / X
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Context only: Gary Marcus / Marcus on AI,
its internal architecture remains problematic.
Why it matters here: The surrounding passage predicts strong initial excitement followed by closer examination revealing remaining problems; the numbered predictions describe continuing errors and unreliable reasoning. This records the comparison text only and does not validate those predictions.
- Publisher
- Gary Marcus / Marcus on AI
- Kind of source
- Secondary summary
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Gary Marcus vs Sam Altman on GPT-o1 aka Strawberry
Almost exactly what I predicted we would see with GPT-4, back on Christmas Day 2022.
- Where
- paragraph 14
- Published
- 2024-09-12
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-gary-marcus-vs-sam-altman-on-gpt-de96b1-a1: Current. Outcome: Supported. 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-gary-marcus-vs-sam-altman-on-gpt-de96b1-a1.