“Similarly, consider pronouns like “himself”: DALL-E does not do well with sentences like “The man is painting a picture of himself”.”
Our reading, the sentence that was checked:
DALL-E does poorly on sentences with reflexive pronouns such as 'The man is painting a picture of himself'.
Assessment: Unresolved Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Three ideas from linguistics that everyone in AI should know, paragraph 54.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-three-ideas-from-linguistics-that-30f93e-a1
We recorded 26 claims from this post. See every claim from this post.
The wording above is the version this site observed on 2026-10-29. The source may have been edited after 2022-06-22. 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
- pronouns like “himself”
- DALL-E does not do well
- The man is painting a picture of himself
Our interpretation
- The ordinary behavioral reading is that the system does not reliably depict the painter as the person represented in the picture.
- The surrounding passage identifies DALL-E Mini, but this sentence says DALL-E. Mini is the provisional contextual reading; evidence about an OpenAI DALL-E version must be kept separate.
- Does not do well describes recurring substantive errors, not an absolute inability or a requirement for flawless facial likeness.
- The present-tense claim is placed in the same June 2022 experimental context, approximated as June 1 through publication on June 22.
Counts as supported if: Representative tests of the identified contemporary model repeatedly fail to depict the painter as the picture's subject for the quoted prompt and closely related reflexive prompts.
Counts as contradicted if: Representative tests of that model reliably depict the painter as the picture's subject across the quoted prompt and related reflexive prompts.
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 statement has a testable behavioral reading, but the located records do not independently settle it. The closest syntactic study is later, coauthored by Marcus, and identifies DALL-E 2. It cannot establish the June result for the provisionally intended Mini model. The unresolved outcome holds under either model reading with the evidence located.
Limitations
- Model identity and the standard for poor performance remain unspecified.
- An unsuccessful picture alone does not isolate a reflexive-pronoun error.
Evidence (0 items)
No evidence item settled this claim by the cutoff.
What would settle it
Identification of the intended model plus independent, representative historical-version outputs for the quoted prompt and related reflexive prompts, with ratings of identity matching and explicit self-portrait controls.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Three ideas from linguistics that everyone in AI should know
Similarly, consider pronouns like “himself”: DALL-E does not do well with sentences like “The man is painting a picture of himself”.
- Where
- paragraph 54
- Published
- 2022-06-22
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-three-ideas-from-linguistics-that-30f93e-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-three-ideas-from-linguistics-that-30f93e-a1.