“OpenAI’s newly-released video generation system Sora seems, like many prior systems, to have been extensively trained on copyrighted materials.”
Our reading:
OpenAI's Sora appears, like many earlier systems, to have been trained extensively on copyrighted material.
Assessment: Can’t be checked
What happened
Not graded on evidence
Outcome: Can’t be checked
An opinion, a recommendation, or too vague to test. Listed, not scored.
Evidence up to . Checked .
Why
Marcus said Sora 'seems' to have been extensively trained on copyrighted materials. This is a hedged inference, which the supplied rules exclude from scoring. Contemporary disclosures and reporting provide relevant context, but grading a categorical training claim would strengthen his wording. This exclusion does not establish that his inference was false or unreasonable.
Caveats
- The contextual evidence is not a training-corpus audit.
- No conclusion is drawn about infringement or fair use.
Evidence (2 items)
Claims like this aren’t scored, so there is no evidence to show.
How we judged it
- Kind of claim
- Observation
- Deadline
- None given. We don’t add one.
Conditions
- seems (their words)
Where we had to interpret
- Marcus presents a tentative inference about the Sora available when he wrote on December 14, 2024.
- Preserving 'seems' means this is not a categorical assertion about the training corpus. Under the supplied rules, this hedged inference is excluded from scoring.
- 'Extensively' supplies no quantity or proportion. No threshold is added.
- 'Copyrighted materials' can include licensed works; the sentence does not itself establish unauthorized use or infringement.
We don’t grade claims like this. We only check the quote and who said it.
What would settle it
Scoring would require a separately stated firm claim. It would not be fair to supply that commitment on Marcus's behalf.
We found 6 claims in this post. See them all.
Full record
Source and dates
Marcus on AI, Generative AI’s Continuing Copyright Problems, an Essay in Memory of Suchir Balaji, 1998 - 2024, paragraph 10.
- Said
- We read the source
- live page, full text, public
- Evidence up to
- Checked
We read this wording on 2026-10-28. The post may have been edited after 2024-12-14.
Where it was said (1)
Said once in the posts we read.
: Marcus on AI, Generative AI’s Continuing Copyright Problems, an Essay in Memory of Suchir Balaji, 1998 - 2024
OpenAI’s newly-released video generation system Sora seems, like many prior systems, to have been extensively trained on copyrighted materials.
- Where
- paragraph 10
- Published
- 2024-12-14
- We read it
- , live page
- Access
- full text, public
History
Versions
- Current: Can’t be checked. Evidence up to 2026-10-28.
What happened to this record
PublishedBy us
Published after both AI reviews and the final check passed.
Held backBy us
A reviewer said our test wasn’t fair to what was actually said.