“His hypothesis, now generally called scaling maximalism, remains extremely popular, in no small part because bigger and bigger models have indeed continued to do ever more impressive things.”
Our reading, the sentence that was checked:
Scaling maximalism remains extremely popular, in no small part because bigger and bigger AI models have continued to do ever more impressive things.
Assessment: Unresolved Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, What “game over” for the latest paradigm in AI might look like, paragraph 9.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-what-game-over-for-the-latest-paradigm-bdf140-a1
We recorded 13 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 2022-10-29. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Causal claim
- 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
- in no small part (qualifier, stated by the speaker)
From the speaker
- remains extremely popular
- in no small part because
- bigger and bigger models
- continued to do ever more impressive things
Our interpretation
- Scaling maximalism means the contextual hypothesis that scaling existing approaches could produce AGI without fundamental new innovation.
- The ordinary reading is widespread enthusiasm in the AI research and industry community, substantially encouraged by demonstrated capability gains. Prominent endorsements alone do not establish its prevalence.
- The context concerns the period from the May 2022 endorsement through publication. No numerical popularity threshold or causal share is supplied.
Counts as supported if: Contemporaneous evidence establishes widespread acceptance of the hypothesis and shows that demonstrated gains from larger models substantially motivated that acceptance.
Counts as contradicted if: Evidence covering the relevant community establishes that the hypothesis had little acceptance, or that capability gains did not substantially motivate its acceptance.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Does the evidence support the cause?
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
Capability gains were celebrated, and Gato's authors explicitly said those gains inspired their work. However, enthusiasm for scaling does not establish widespread belief that it suffices for AGI. The strongest prevalence evidence found points the other way: only 17% of surveyed NLP researchers endorsed a related maximalist proposition. That survey covers a narrower community and a different question. The full popularity-and-causation claim remains unsettled.
Limitations
- The survey concerns NLP researchers, not all AI researchers, companies, investors, or the public.
- Research motivation is not equivalent to belief that scaling alone can deliver AGI.
Evidence (3 items)
Context only: Julian Michael and colleagues, arXiv,
only a small minority (17%) actually agree with the position
Why it matters here: The May–June survey found only 17% agreement that scaling established techniques could practically solve any important NLP problem. This challenges broad popularity, but its population and proposition do not exactly match Marcus's unspecified AI community and AGI hypothesis.
- Publisher
- Julian Michael and colleagues, arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Context only: Scott Reed and colleagues, arXiv,
Inspired by progress in large-scale language modeling
Why it matters here: The researchers explicitly connect prior capability gains to their generalist-agent research. This supports the proposed motivation in one important example, but does not establish widespread belief that scaling alone suffices for AGI.
- Publisher
- Scott Reed and colleagues, arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Context only: Google Research,
achieved impressive results across a wide range of tasks.
Why it matters here: This establishes contemporary enthusiasm about demonstrated capabilities before the period discussed. It measures neither maximalism's prevalence nor the causes of that prevalence.
- Where
- Opening paragraph
- Publisher
- Google Research
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
What would settle it
A contemporary survey or comparable systematic record covering the intended community, distinguishing scaling's usefulness from its sufficiency for AGI, and documenting why adherents held that belief.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, What “game over” for the latest paradigm in AI might look like
His hypothesis, now generally called scaling maximalism, remains extremely popular, in no small part because bigger and bigger models have indeed continued to do ever more impressive things.
- Where
- paragraph 9
- Published
- 2022-10-29
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-what-game-over-for-the-latest-paradigm-bdf140-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-what-game-over-for-the-latest-paradigm-bdf140-a1.