“You can read the full paper by Timothy Duggan, Pierrick Lorang, Hong Lu, and Matthias Scheutz, which very much vindicates the Apple reasoning paper, here.”
Our reading, the sentence that was checked:
The Tufts paper strongly vindicates the Apple reasoning paper's findings.
Assessment: Unresolved Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Even more good news for the future of neurosymbolic AI, paragraph 17.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-even-more-good-news-for-the-future-d1a541-a1
We recorded 11 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 2026-04-12. 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 full paper by Timothy Duggan, Pierrick Lorang, Hong Lu, and Matthias Scheutz
- very much vindicates the Apple reasoning paper
Our interpretation
- Read vindicates as independently corroborates Apple's finding of planning and generalization failures as task complexity increases.
- The surrounding passage identifies a conceptual extension to VLAs, not a direct replication using Apple's models.
- Strong corroboration requires more than similar task-completion scores if perception or movement failures can explain the difference.
- Assess the Tufts findings reported in February 2026; Apple's earlier findings are comparison context.
Counts as supported if: The Tufts methods and results independently demonstrate a comparable complexity-linked planning or reasoning failure, with execution confounds sufficiently separated.
Counts as contradicted if: The Tufts results directly overturn the relevant Apple finding or clearly demonstrate that the claimed corresponding reasoning failure is absent.
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 studies show compatible performance limitations, but compatibility does not settle strong vindication. The execution confound prevents the Tufts result from independently establishing the same reasoning failure. It also does not overturn Apple's finding.
Limitations
- A weaker reading of vindicates as merely consistent with would receive more support.
- This assessment concerns the claimed relationship between the studies, not whether Apple's findings are independently correct.
Evidence (2 items)
Context only: arXiv; Apple researchers,
Through extensive experimentation across diverse puzzles, we show that frontier LRMs face a complete accuracy collapse beyond certain complexities.
Why it matters here: Apple reports complexity-linked collapse in language-model puzzle solving. This establishes the comparison target, not independent confirmation of it.
- Where
- Abstract
- Publisher
- arXiv; Apple researchers
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Context only: arXiv; Duggan, Lorang, Lu and Scheutz,
Their task advancement rates were also very low.
Why it matters here: The task failures resemble Apple's pattern, but the discussion attributes failures mainly to low-level execution. The planner-guided VLA failed despite receiving valid plans.
- Publisher
- arXiv; Duggan, Lorang, Lu and Scheutz
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
What would settle it
A comparison that separates planning correctness from perception and movement, then tests whether the same complexity-linked reasoning failure remains.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Even more good news for the future of neurosymbolic AI
You can read the full paper by Timothy Duggan, Pierrick Lorang, Hong Lu, and Matthias Scheutz, which very much vindicates the Apple reasoning paper, here.
- Where
- paragraph 17
- Published
- 2026-04-12
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-even-more-good-news-for-the-future-d1a541-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-even-more-good-news-for-the-future-d1a541-a1.