“LLMs are not AGI, and (on their own) never will be; scaling alone was never going to be enough.”
Our reading, the sentence that was checked:
Scaling alone is not enough to reach AGI.
Assessment: Unresolved Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, The Great AI Retrenchment has begun, paragraph 4.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-the-great-ai-retrenchment-has-begun-b90353-2-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 2024-06-15. 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
- scaling alone (qualifier, stated by the speaker)
From the speaker
- scaling alone
- was never going to be enough
Our interpretation
- Enough means sufficient to reach AGI, as established by the preceding clause.
- Scaling alone means increasing model size, training data and training compute while retaining the underlying LLM method.
- Broad human-level cognitive competence is the ordinary reading of AGI.
- New learning methods, external tools or other substantive system changes must be distinguished from increases in scale.
Counts as supported if: A general theoretical or causal result establishes that increasing scale within the specified method cannot reach broad human-level cognitive competence.
Counts as contradicted if: A system reaches independently established AGI through scaling the specified method alone, with evidence separating scale from substantive methodological changes.
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
Scaling research demonstrates improvements in particular metrics, while limitations research identifies constraints. Neither establishes whether scaling alone can reach AGI. Recent model disclosures also combine several research methods, so their results do not isolate scaling's contribution. The causal claim remains unsettled.
Limitations
- Finite scaling experiments cannot establish what every larger system could achieve.
- A limit on perfect accuracy is not necessarily a limit below human competence.
- The boundary between scaling and methodological change is unspecified.
Evidence (3 items)
Context only: arXiv; Kaplan and colleagues,
the cross-entropy loss
Why it matters here: The paper studies predictable improvements in a language-model training metric. That outcome does not establish whether scaling suffices for AGI.
- Publisher
- arXiv; Kaplan and colleagues
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Context only: arXiv; Mohsin and colleagues,
theoretical ceilings
Why it matters here: The paper presents limitations involving error, context and reasoning. The existence of such limitations does not establish that scaled LLMs cannot attain human-level competence, which also permits errors and limits.
- Publisher
- arXiv; Mohsin and colleagues
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Context only: OpenAI,
pre-training, reinforcement learning
Why it matters here: The developer describes several ingredients behind this model. This disclosure does not isolate scaling as the cause of its capabilities or establish that scaling alone is insufficient.
- Publisher
- OpenAI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
What would settle it
A documented scaling-only system independently shown to reach AGI, or a validated causal or theoretical result showing that the specified scaling process cannot reach human-level general competence.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, The Great AI Retrenchment has begun
LLMs are not AGI, and (on their own) never will be; scaling alone was never going to be enough.
- Where
- paragraph 4
- Published
- 2024-06-15
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-the-great-ai-retrenchment-has-begun-b90353-2-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-the-great-ai-retrenchment-has-begun-b90353-2-a1.