Pundit ClaimsAI claims by Gary Marcus and Ed Zitron, with the evidence

“But OpenAI doesn’t want to do anything nearly as conventional as that. They want to continue to use copyrighted materials, apparently without paying for them.”

DALL-E’s New Guardrails: Fast, Furious, and Far from Airtight, paragraph 3. The full statement is at the source.

Our reading, the sentence that was checked:

OpenAI does not want to take approaches as conventional as training only on public domain data or licensing other data, and instead wants to keep using copyrighted materials, apparently without paying for them.

Assessment: Can’t be checked Evidence as of

Where and when Gary Marcus wrote it

Marcus on AI, DALL-E’s New Guardrails: Fast, Furious, and Far from Airtight, paragraph 3.

Said
Source observed
live page, full text, public
Evidence as of
the cutoff for this assessment
Assessment recorded
assessment marcus-dall-es-new-guardrails-fast-furious-37170d-a2

We recorded 9 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 2024-01-10. 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

  • apparently (applies to 'without paying for them') (qualifier, stated by the speaker)

From the speaker

  • doesn’t want to do anything nearly as conventional as that
  • continue to use copyrighted materials
  • apparently without paying for them
  • train only on public domain training data
  • license any other data

Our interpretation

  • The preceding sentences describe comprehensive alternatives: public-domain-only training or licensing the remaining material. Rejecting those alternatives does not mean rejecting every licensing agreement.
  • The firm component concerns OpenAI’s publicly expressed training policy. Public statements can establish that position without establishing private motives.
  • The payment allegation is expressly tentative. 'Apparently' must not become a categorical assertion that OpenAI paid nobody, or a firm commitment about some evaluator-selected proportion of its data.
  • The relevant period is January 8–10, 2024: the response discussed in the post through the post date. This is an ongoing policy claim, not a quarterly financial observation.
  • The compound claim cannot receive a firm truth verdict without supplying a commitment that the payment qualification leaves unspecified. Its firm policy component can nevertheless be examined.
  • This is a qualitative claim about copyright and training policy, not a claim specifying a financial metric.

This claim is not graded on evidence. The record checks that the words are quoted correctly and attributed to the right person, and that the claim is classified correctly.

Current assessment

Not graded on evidence

Outcome: Can’t be checked

The statement is a recommendation, a value judgment, or too open to test. It is recorded, not scored.

Evidence as of
Deadline state when assessed
No horizon
Assessment recorded
Policy version
2026-10-28.7

Rationale

The firm policy component is checkable: OpenAI publicly defended training on internet material under fair use. AP also reported that OpenAI paid to license its archive. Selective licensing does not refute rejection of an exclusively public-domain-or-licensed approach. But Marcus expressly qualified the payment allegation with 'apparently'. A firm verdict on this compound claim would either remove that qualification or supply a scope and confidence threshold he did not state. The evidence therefore informs the assessment without supporting a binary grade for the claim as submitted.

Limitations

  • This does not reject the independently checkable policy component.
  • OpenAI’s fair-use assertion is evidence of its position, not a finding that its conduct was lawful.
  • The syndicated AP version appeared on the post date; its publication time relative to Marcus’s post is not established.

Evidence (2 items)

No outcome evidence: a claim of this kind is recorded and attributed, not scored.

What would settle it

Separately assess the firm policy component while retaining the payment sentence as a tentative inference, or obtain clarification of the payment allegation’s intended scope and commitment.

Where it was said (1)

Said once in the sources this register covers.

  • : Marcus on AI, DALL-E’s New Guardrails: Fast, Furious, and Far from Airtight

    But OpenAI doesn’t want to do anything nearly as conventional as that. They want to continue to use copyrighted materials, apparently without paying for them.

    Published
    2024-01-10
    Source observed
    , live page
    Access
    full text, public

History

Assessment versions

Recorded events

  1. Assessment publishedBy this site

    First assessment published after both reviews and the final-text check passed.

    About assessment marcus-dall-es-new-guardrails-fast-furious-37170d-a2.

  2. HeldBy this site

    The openai review found the criteria unfair to the speaker's words.

    About assessment marcus-dall-es-new-guardrails-fast-furious-37170d-a1, which was not published.