“Yet nearly all funding has been aimed simply at making them larger.”
Our reading, the sentence that was checked:
Nearly all AI funding has gone to making LLMs larger.
Assessment: Unresolved Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, How o3 and Grok 4 Accidentally Vindicated Neurosymbolic AI, paragraph 47.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-how-o3-and-grok-4-accidentally-vindicated-e347cb-a1
We recorded 36 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 2025-07-13. 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
- nearly all funding
- aimed simply at making them larger
Our interpretation
- Them refers to LLMs. The ordinary reading is an overwhelming share of AI funding directed solely toward increasing LLM scale.
- Funding a company that develops LLMs does not establish that its money was devoted solely to larger models.
- Use calendar 2024 as the latest complete annual funding period publicly reportable before the post. The alternative cumulative reading has no specified starting date.
- This is an aggregate sector allocation claim, not a financial figure for a named company; metric_scope is null.
Counts as supported if: Comparable funding and allocation records show that an overwhelming share of AI funding was directed solely toward increasing LLM size or pretraining scale.
Counts as contradicted if: Comparable records establish that a substantial share of the same funding total was allocated to other purposes.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Was it true for the period described?
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
The searched records do not measure the asserted funding share. Investment totals identify recipients or sectors, while the more detailed compute analysis covers particular companies and different spending categories. It cannot establish what fraction of all AI funding pursued LLM enlargement. The available evidence therefore neither supports nor decisively contradicts nearly all.
Limitations
- The selected annual period cannot settle an unspecified cumulative historical claim.
- The retrospective compute analysis was published after the post but before the evaluation cutoff.
- No numerical threshold or accounting basis can fairly be supplied on the speaker's behalf.
Evidence (1 item)
Context only: Epoch AI,
companies rarely disclose
Why it matters here: Epoch estimates that roughly $500 million of OpenAI's approximately $5 billion in 2024 R&D compute went to final training runs. This challenges equating all compute expenditure with final training, but experiments and other research may also pursue scale, and one company's compute spending does not measure all AI funding.
- Publisher
- Epoch AI
- Kind of source
- Independent measurement
- Measures
- OpenAI, R&D compute expenditure and its allocation to final training runs, USD billions, estimate, 2024-01-01 to 2024-12-31
- Describes events of
- Public from
What would settle it
A defined funding universe and comparable allocation records separating LLM enlargement from other purposes, with enough coverage to determine whether the former overwhelmingly dominated.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, How o3 and Grok 4 Accidentally Vindicated Neurosymbolic AI
Yet nearly all funding has been aimed simply at making them larger.
- Where
- paragraph 47
- Published
- 2025-07-13
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-how-o3-and-grok-4-accidentally-vindicated-e347cb-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-how-o3-and-grok-4-accidentally-vindicated-e347cb-a1.