“Computers can answer some questions reliably now, for example, and some not, and the graph tells us nothing about which is which or when any specific question will be solved.”
Our reading, the sentence that was checked:
The METR graph tells us nothing about which questions computers can reliably answer or when specific questions will be solved.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, The latest AI scaling graph - and why it hardly makes sense, paragraph 18.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-the-latest-ai-scaling-graph-and-why-f8db9d-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 2025-05-03. 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
- the graph tells us nothing
- which is which
- when any specific question will be solved
Our interpretation
- Assess the March graph and April update discussed in the post.
- The ordinary contextual reading is that the graph does not identify specific reliably answerable questions or forecast their individual solution dates.
- Do not read nothing as denying that an aggregate statistical relationship provides any information at all.
- Assess the graph’s defined output, rather than treating its underlying task-level dataset as the graph itself.
Counts as supported if: The graph reports aggregate task-duration success thresholds over model release dates, without identifying specific reliably solved questions or assigning dates to their future solution.
Counts as contradicted if: The graph provides validated question-specific reliability classifications or forecasts when identified questions will become solvable.
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 graph summarizes task-duration success thresholds and their change across model releases. It does not identify which particular questions computers reliably answer or predict when particular questions will be solved. This supports the ordinary question-specific reading. A stronger reading that the graph conveys no useful statistical information would not follow from this evidence.
Limitations
- The outcome depends on reading the claim as about specific questions.
- The underlying study contains task-level observations and statistical relationships; those are not question-specific forecasts supplied by the graph.
- This assessment does not establish that the aggregate metric is useless.
Evidence (1 item)
Supports the claim: METR researchers via arXiv,
This is the time humans typically take to complete tasks that AI models can complete with 50% success rate.
Why it matters here: The defined output is a duration associated with aggregate success probability. The paper’s fitting and extrapolation methods do not turn this into a list of reliably answered questions or dates for solving identified questions.
- Publisher
- METR researchers via arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, The latest AI scaling graph - and why it hardly makes sense
Computers can answer some questions reliably now, for example, and some not, and the graph tells us nothing about which is which or when any specific question will be solved.
- Where
- paragraph 18
- Published
- 2025-05-03
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-the-latest-ai-scaling-graph-and-why-f8db9d-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-the-latest-ai-scaling-graph-and-why-f8db9d-a1.