“Google and Anthropic have each largely caught up; various Chinese companies are closing in.”
Our reading, the sentence that was checked:
Google and Anthropic have largely caught up with OpenAI, and Chinese companies are closing in.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Breaking: OpenAI is probably toast, paragraph 1.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-breaking-openai-is-probably-toast-d0cad9-a1
We recorded 6 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 2026-02-13. 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
- Google and Anthropic have each largely caught up
- various Chinese companies are closing in
Our interpretation
- The ordinary reading in this context concerns frontier model capabilities, not company revenue, market share, or profitability.
- 'Largely caught up' means broadly competitive performance across important capability evaluations; it does not require winning every task.
- 'Various Chinese companies' requires evidence concerning more than one Chinese developer.
- 'Closing in' requires improvement toward the frontier, rather than merely the existence of competing models.
- The current comparison is assessed using evaluations public by February 13, 2026. The target period includes late 2025 because the statement also describes a recent direction of change.
Counts as supported if: Contemporaneous evaluations place Google and Anthropic alongside or ahead of OpenAI on important capabilities, while evaluations of at least two Chinese developers show meaningful progress toward that frontier.
Counts as contradicted if: Comparable evaluations show a persistent broad capability separation favoring OpenAI over Google or Anthropic, or show Chinese competitors collectively falling farther behind rather than approaching the frontier.
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
The capability reading is supported. Google led the cited professional-agent evaluation, and Anthropic led Artificial Analysis's aggregate index by February 6. Independent reports also documented frontier-oriented gains by Moonshot and Z.ai. These results support broad competitiveness and Chinese progress, rather than equality on every task.
Limitations
- The result depends on interpreting the statement as a claim about model capabilities.
- Three items come from one independent evaluator.
- Benchmark results depend on model settings and task selection.
- Google's cited independent comparison is from January, while the Anthropic and GLM results are from February.
Evidence (4 items)
Supports the claim: arXiv; Vidgen and colleagues,
highest score
Why it matters here: The original APEX-Agents evaluation places a Google model ahead of OpenAI on professional, multi-application tasks. This supports competitive capability in an important domain, although it is not a universal ranking.
- Publisher
- arXiv; Vidgen and colleagues
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Supports the claim: Artificial Analysis,
achieves the highest score
Why it matters here: The report places Opus 4.6 first on its ten-evaluation Intelligence Index, ahead of the previous leader GPT-5.2, and identifies leadership in agentic work, terminal coding, and physics. This supports Anthropic having caught up while preserving the report's task-specific qualifications.
- Publisher
- Artificial Analysis
- Kind of source
- Independent measurement
- Describes events of
- Public from
Supports the claim: Artificial Analysis,
leaps ahead
Why it matters here: The report describes Moonshot's model as closer to the frontier and places it behind only OpenAI and Anthropic on GDPval-AA. It also reports visual reasoning performance comparable to GPT-5.2 and Opus 4.5.
- Publisher
- Artificial Analysis
- Kind of source
- Independent measurement
- Describes events of
- Public from
Supports the claim: Artificial Analysis,
the new leading open weights model
Why it matters here: The report measures an Intelligence Index rise from GLM-4.7's 42 to GLM-5's 50 and explicitly describes a narrowing gap. Its agentic ranking, behind Opus 4.6 and GPT-5.2, supports progress by a second Chinese developer without claiming complete parity.
- Where
- Opening summary
- Publisher
- Artificial Analysis
- 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, Breaking: OpenAI is probably toast
Google and Anthropic have each largely caught up; various Chinese companies are closing in.
- Where
- paragraph 1
- Published
- 2026-02-13
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-breaking-openai-is-probably-toast-d0cad9-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-breaking-openai-is-probably-toast-d0cad9-a1.