“People had grown to expect miracles, but GPT-5 is just the latest incremental advance.”
Our reading, the sentence that was checked:
GPT-5 is merely an incremental advance, not a radical leap.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, GPT-5: Overdue, overhyped and underwhelming. And that’s not the worst of it., paragraph 29.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-gpt-5-overdue-overhyped-and-underwhelming-72c33c-a1
We recorded 23 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 2025-08-09. 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
- People had grown to expect miracles
- GPT-5 is just the latest incremental advance
Our interpretation
- Read as: at launch, GPT-5 was a modest step over the frontier models just before it (o3, Grok 4), not a large jump. The surrounding text ('not that different from anything that came before', 'barely better than... Grok 4') supports this reading.
- Target period is August 2025.
Counts as supported if: Launch-era evaluations show GPT-5 only modestly ahead of the immediately preceding frontier models.
Counts as contradicted if: Launch-era evaluations show GPT-5 far ahead of o3 and other then-current frontier models, a step change of the kind seen from GPT-3 to GPT-4.
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
In context, Marcus calls GPT-5 'not that different from anything that came before' and 'barely better' than Grok 4. So 'the latest incremental advance' reads as a small step over the models just before it. At launch, Artificial Analysis said GPT-5's increase over o3 was 'not comparable to the jump from GPT-3 to GPT-4', and its index showed GPT-5 only one or two points above o3 and Grok 4. That supports the claim. Against GPT-4, though, Epoch AI found GPT-5 to be a large leap, similar to GPT-3 to GPT-4, with the gains spread across many intermediate releases. Under that reading the claim would fail.
Limitations
- The result depends on reading the comparison as against the immediately preceding models.
- The Artificial Analysis assessment comes via a news report quoting it.
- 'Incremental' is a judgment, not a fixed threshold.
Evidence (3 items)
Supports the claim: OfficeChai (quoting Artificial Analysis),
While High sets a new standard, the increase over o3 is not comparable to the jump from GPT-3 to GPT-4 or GPT-4o to o1,” Artificial Analysis said.
Why it matters here: An independent benchmark group judged GPT-5's gain over o3 to be small compared with earlier big jumps. That matches 'incremental'.
- Where
- paragraph 3
- Publisher
- OfficeChai (quoting Artificial Analysis)
- Kind of source
- News report
- Describes events of
- Public from
Supports the claim: OfficeChai,
GPT-5 barely managed to push the frontier from o3 and Grok 4.
Why it matters here: The report's own reading of the launch index: only a small gain over the prior frontier.
- Where
- final paragraph
- Publisher
- OfficeChai
- Kind of source
- News report
- Describes events of
- Public from
Context only: Epoch AI,
Yet GPT-5 does far better than GPT-4 in some notable capability benchmarks, similar to how GPT-4 advanced on GPT-3 on the widely-cited benchmarks of its day.
Why it matters here: Measured against GPT-4, GPT-5 was a large leap. Epoch adds that many intermediate releases spread out the gains. This cuts against the claim if 'incremental' is read as relative to GPT-4.
- Where
- paragraph 1
- Publisher
- Epoch AI
- Kind of source
- Independent measurement
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, GPT-5: Overdue, overhyped and underwhelming. And that’s not the worst of it.
People had grown to expect miracles, but GPT-5 is just the latest incremental advance.
- Where
- paragraph 29
- Published
- 2025-08-09
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-gpt-5-overdue-overhyped-and-underwhelming-72c33c-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-gpt-5-overdue-overhyped-and-underwhelming-72c33c-a1.