“The idea that we have reached that point is laughable.”
Our reading, the sentence that was checked:
Current AI has not reached a level beyond AGI that can do anything an expert human can and much more.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Sorry, Sam and Elon, we have not reached the Singularity, paragraph 72.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-sorry-sam-and-elon-we-have-not-reached-fc6d75-a1
We recorded 17 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 2026-07-28. 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)
Conditions
- I. J. Good sense: beyond AGI that can do anything an expert can (qualifier, stated by the speaker)
From the speaker
- a step beyond AGI
- anything a person, even an expert, can do, and much more
- The idea that we have reached that point is laughable
Our interpretation
- Assess the factual assertion that current AI had not surpassed expert humans across intellectual activities. Laughable is rhetorical emphasis, not a separate measurable claim.
- The contextual definition supplies the capability standard, not a conditional forecast.
- The plain reading requires reliable breadth of capability. Strong performance in selected domains alone is insufficient.
- Use July 1–28, 2026, for current capabilities. There is no regular reporting period for frontier capability.
- The claim does not require independently grading the separate assertion about how many Marcus–Brundage tasks AI could perform.
Counts as supported if: Credible July evaluations show that leading AI systems still have substantial deficits on intellectual tasks humans can perform.
Counts as contradicted if: Broad, independently validated evaluations establish an AI system that reliably matches or surpasses expert humans across intellectual activities, including unfamiliar tasks.
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
July evaluations identify a substantial gap on unfamiliar reasoning tasks that humans can solve. The July 24 benchmark leader still scored 30.2% on ARC-AGI-3. These results support the claim that publicly evaluated AI had not achieved reliable superiority across intellectual activities. High scores on other benchmarks do not remove the demonstrated gap.
Limitations
- ARC-AGI-3 measures a particular kind of adaptive reasoning, not intelligence as a whole.
- The July 24 ARC-AGI-3 evaluation used High reasoning effort because the testing window was short.
- The evidence concerns publicly evaluated systems and does not verify the separate Marcus–Brundage task count.
Evidence (2 items)
Supports the claim: ARC Prize,
remains hard for today's AI.
Why it matters here: The benchmark provider describes a contemporary gap on unfamiliar interactive reasoning tasks humans can solve. That gap is inconsistent with reliable superiority across intellectual activities.
- Where
- Opening paragraph
- Publisher
- ARC Prize
- Kind of source
- Benchmark result
- Describes events of
- Public from
Supports the claim: ARC Prize,
scoring 30.2%.
Why it matters here: The dated result shows substantial remaining limitations even for the benchmark leader shortly before the post. The page also records strong results on earlier benchmarks, but those do not establish superiority across all intellectual activities.
- Publisher
- ARC Prize
- Kind of source
- Benchmark result
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Sorry, Sam and Elon, we have not reached the Singularity
The idea that we have reached that point is laughable.
- Where
- paragraph 72
- Published
- 2026-07-28
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-sorry-sam-and-elon-we-have-not-reached-fc6d75-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-sorry-sam-and-elon-we-have-not-reached-fc6d75-a1.