“It can see (though we don’t have a lot of details on that yet); it does astonishingly well on a whole bunch of standardized tests, like LSATs, GREs, and SATs.”
Our reading, the sentence that was checked:
GPT-4 has vision capability and performs astonishingly well on standardized tests such as the LSAT, GRE, and SAT.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, GPT-4’s successes, and GPT-4’s failures, paragraph 2.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-gpt-4s-successes-and-gpt-4s-failures-ce06c5-a2
We recorded 12 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 2023-03-15. 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)
From the speaker
- It can see
- astonishingly well
- standardized tests, like LSATs, GREs, and SATs
Our interpretation
- Seeing means accepting images and answering questions about their contents, without implying human-like visual understanding.
- The ordinary reading of strong test performance is high standing relative to human test takers across the named examinations, not exceptional performance on every subsection.
- The target is the original GPT-4 at publication. The launch-to-post interval is the latest publicly documented snapshot of this ongoing capability.
Counts as supported if: Launch-model documentation establishes image understanding and high reported performance across the LSAT, GRE and SAT.
Counts as contradicted if: The launch model lacks image understanding, or comparable results show generally ordinary or poor performance across the named examinations.
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
OpenAI's original report, available March 15, 2023, documents image-input capability in launch-era GPT-4 [e1]. It reports approximately the 88th percentile on LSAT, 93rd on SAT reading/writing, 89th on SAT mathematics, 80th on GRE quantitative and 99th on GRE verbal [e1]. GRE writing was weaker, at approximately the 54th percentile [e1]. This supports strong reported performance across the named examinations, without establishing exceptional performance on every subsection [e1].
Limitations
- The evidence is OpenAI's own evaluation report, without independent replication in the packet [e1].
- The reported evaluations do not establish scores from officially administered candidate sittings.
- Image-input capability alone does not establish the scope of public access at launch.
Evidence (1 item)
Supports the claim: OpenAI, published on arXiv,
This technical report presents GPT-4, a large multimodal model capable of processing image and text inputs and producing text outputs.
Why it matters here: The report documents launch-model image processing and reports approximately the 88th percentile on LSAT, 93rd on SAT reading/writing, 89th on SAT mathematics, 80th on GRE quantitative and 99th on GRE verbal. GRE writing was approximately the 54th percentile; event_date denotes the launch model's capability, not an asserted examination date.
- Publisher
- OpenAI, published on arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, GPT-4’s successes, and GPT-4’s failures
It can see (though we don’t have a lot of details on that yet); it does astonishingly well on a whole bunch of standardized tests, like LSATs, GREs, and SATs.
- Where
- paragraph 2
- Published
- 2023-03-15
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-gpt-4s-successes-and-gpt-4s-failures-ce06c5-a2: 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-gpt-4s-successes-and-gpt-4s-failures-ce06c5-a2.