“Dall-E already has one minor safeguard in place – proper names (and hence deliberate infringement attempts) reportedly sometimes get blocked – but those safeguards aren’t fully reliable:”
Our reading:
DALL-E's proper-name blocking safeguards against infringement are not fully reliable.
Assessment: Unresolved
What happened
Was it true for the period described?
Outcome: Unresolved
We checked, and the evidence so far doesn’t settle it.
Judged on a plain reading. They named no measure, so we used the reading a typical informed reader would take. It is spelled out under “How we judged it”.
Evidence up to . Checked .
Why
On November 20, 2023, The Register reported that Bing's temporary block on “Disney” had been lifted [e2]. It also reported that Bing no longer reproduced Disney's wording or logo [e2]. This November account does not establish whether an intended name restriction failed in Bing during December 1–29, 2023, leaving the claim unresolved [e2].
Caveats
- No verified December record establishes both the relevant Bing safeguard's intended coverage and a failure within that coverage.
- The reported November improvement does not establish full reliability or December behavior.
- The packet lacks deployment records and comprehensive testing that could rule out the claimed failures.
Evidence (1 item)
Context only: The Register,
reportedly doesn't recreate
Why it matters: This news report describes a temporary Disney-name ban followed by an output restriction. It shows that the safeguard changed and reportedly worked for that case, but does not establish its December reliability.
- Publisher
- The Register
- Kind of source
- News report
- About events on
- Published
How we judged it
- Kind of claim
- Observation
- Deadline
- None given. We don’t add one.
Where we had to interpret
- In context, this concerns DALL-E accessed through Bing, rather than assuming that Bing and ChatGPT had identical safeguards.
- I use December 1–29, 2023 as the latest contemporaneously observable month before publication. The post supplies no exact experiment dates.
- The ordinary reading is that a proper-name safeguard existed but could fail to prevent the targeted character or work from being generated.
- Acceptance of a name outside the safeguard's intended coverage would not, by itself, establish a failure.
Right if: Contemporary records establish the relevant name-blocking safeguard and document at least one failure or bypass involving a name or work it was intended to restrict.
Wrong if: Authoritative deployment records and sufficiently comprehensive testing establish that the claimed failures did not occur, or that the purported safeguard did not exist in the specified deployment.
We wrote this test after the outcome was known.
What would settle it
December 1–29, 2023 Bing records showing the safeguard's intended coverage and a failure within it, with dated prompts and results; or authoritative deployment records and comprehensive testing establishing that the claimed failures did not occur or the purported safeguard did not exist.
We found 16 claims in this post. See them all.
Full record
Source and dates
Marcus on AI, Things are about to get a lot worse for Generative AI, paragraph 6.
- Said
- We read the source
- live page, full text, public
- Evidence up to
- Checked
We read this wording on 2026-10-29. The post may have been edited after 2023-12-29.
Where it was said (1)
Said once in the posts we read.
: Marcus on AI, Things are about to get a lot worse for Generative AI
Dall-E already has one minor safeguard in place – proper names (and hence deliberate infringement attempts) reportedly sometimes get blocked – but those safeguards aren’t fully reliable:
- Where
- paragraph 6
- Published
- 2023-12-29
- We read it
- , live page
- Access
- full text, public
History
Versions
- Current: Unresolved. 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
The final check failed: a cited source didn’t say what the write-up said.