“Turns out Sora makes stuff up, too,”
Our reading, the sentence that was checked:
OpenAI's Sora, like LLMs, makes things up.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Sora Can’t Handle the Truth, paragraph 2.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-sora-cant-handle-the-truth-a90f3d-a2
We recorded 11 claims from this post. See every claim from this post.
The wording above is the version this site observed on 2026-10-29. The source may have been edited after 2024-02-17. 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
- Sora makes stuff up, too
- the chessboard itself
- rendered the image photorealistically
Our interpretation
- The plain reading is that Sora can produce realistic-looking depictions containing structural or physical errors. Merely generating an imaginary scene does not count.
- The target is the original Sora demonstrated around the post, using the February 15–17, 2024 launch window as the latest publicly observable period.
- The comparison with LLMs concerns the existence of errors, not equal error rates or identical mechanisms.
Counts as supported if: An authenticated launch-era Sora output or contemporaneous developer disclosure establishes an unintended structural or physical error.
Counts as contradicted if: Evidence establishes that the alleged examples were misattributed or were requested fictional features, and that the relevant outputs did not contain the claimed class of error.
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
OpenAI’s February 15, 2024 launch disclosure labels physically implausible motion in a Sora running demonstration as a weakness [e1]. This supports the contextual claim that launch-era Sora produced physical errors, beyond merely generating requested imaginary scenes [e1].
Limitations
- The disclosure establishes the occurrence of errors, not their frequency or how convincing viewers find the outputs.
- It does not verify the chessboard dimensions or piece counts.
- It does not establish equal error rates or shared error mechanisms between Sora and LLMs.
Evidence (1 item)
Supports the claim: OpenAI,
Sora sometimes creates physically implausible motion.
Why it matters here: OpenAI identifies an unintended physical error in a launch demonstration. This supports the existence of realistic-looking errors without depending on Marcus's chessboard analysis.
- Publisher
- OpenAI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Sora Can’t Handle the Truth
Turns out Sora makes stuff up, too,
- Where
- paragraph 2
- Published
- 2024-02-17
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-sora-cant-handle-the-truth-a90f3d-a2: 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-sora-cant-handle-the-truth-a90f3d-a2.