“This is in line with the July METR study showing that coding tools sometimes impair productivity rather than increasing it.”
Our reading, the sentence that was checked:
The July METR study showed that AI coding tools sometimes impair productivity rather than increase it.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Two takes on AI and the future of America, paragraph 3.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-two-takes-on-ai-and-the-future-of-a11d10-a1
We recorded 8 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 2025-12-23. 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
- according to the July METR study (qualifier, stated by the speaker)
- sometimes (qualifier, stated by the speaker)
From the speaker
- the July METR study
- coding tools sometimes impair productivity rather than increasing it
Our interpretation
- Assess what the July 2025 publication found, rather than the performance of every coding tool.
- Sometimes means at least the circumstances studied; it does not imply a typical effect across all developers.
- The target period is July 2025 because the claim concerns that publication's findings. Its experiments used earlier-2025 tools.
Counts as supported if: The July METR study reports that allowing AI coding tools increased task-completion time in the studied setting.
Counts as contradicted if: The study reports no slowdown, or its design and results do not support the claimed existence of a productivity impairment.
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 July METR paper explicitly reports that AI access increased completion time by 19% in its randomized trial. Marcus's word 'sometimes' preserves the study's limited scope. Later evidence about improved tools does not reverse what that study found.
Limitations
- The trial involved experienced developers working on familiar open-source projects.
- The finding does not establish the average effect across all software development.
- The event date refers to the July publication being assessed, not the earlier experimental work.
Evidence (2 items)
Supports the claim: METR authors via arXiv,
increases completion time by 19%
Why it matters here: The July paper reports increased completion time in a randomized trial involving 16 experienced developers and 246 tasks. This directly supports the limited word 'sometimes.'
- Where
- Abstract, version 2
- Publisher
- METR authors via arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Context only: METR,
evidence for speedup
Why it matters here: The later report describes possible speedups but also serious selection problems. It concerns different measurements and does not contradict the historical statement about July's finding.
- Publisher
- METR
- 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, Two takes on AI and the future of America
This is in line with the July METR study showing that coding tools sometimes impair productivity rather than increasing it.
- Where
- paragraph 3
- Published
- 2025-12-23
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-two-takes-on-ai-and-the-future-of-a11d10-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-two-takes-on-ai-and-the-future-of-a11d10-a1.