“To begin with, the threat is empty: open source LLMs are already out in the wild, and a large number of people know how to build them.”
Our reading, the sentence that was checked:
OpenAI's threat that generative AI would disappear is empty because open source LLMs already exist and many people can build them.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, The desperate race to save Generative AI, paragraph 9.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-the-desperate-race-to-save-generative-b5da50-a1
We recorded 13 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 2024-01-08. 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
- the threat is empty
- open source LLMs are already out in the wild
- a large number of people know how to build them
Our interpretation
- The ordinary technical reading is that independently usable LLM releases and substantial development expertise already existed outside OpenAI.
- Open source is read here as publicly downloadable models with permission for independent use, rather than a guarantee that every training input is available.
- A large number means a substantial development community, without inventing a numerical threshold.
- The word already describes accumulated releases and expertise before the post, so the target period has no specified beginning and ends on 2024-01-08.
- The empty-threat inference concerns technical dependence on individual suppliers. It does not establish that OpenAI actually made the attributed threat or that every model was legally immune from copyright claims.
Counts as supported if: Pre-post records show independently usable public LLM releases and a substantial community participating in building and documenting them.
Counts as contradicted if: Reliable pre-post records show that usable LLMs and the expertise to build them remained confined to the commercial suppliers whose withdrawal supposedly threatened availability.
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
Mistral had released a model for independent use before the post, and the BLOOM research report documents a substantial development community. These records support Marcus’s technical point that usable LLMs and relevant knowledge were already distributed beyond OpenAI. They do not prove that OpenAI made the alleged threat or that copyright restrictions could have no effect on those models.
Limitations
- The result uses the ordinary public-model reading of open source.
- Project contributors had different roles; their number is not a count of people able to train a complete model alone.
- Technical availability does not establish legal immunity, affordability, or indefinite persistence.
Evidence (2 items)
Supports the claim: Mistral AI,
under the Apache 2.0 license
Why it matters here: The announcement establishes a pre-post public model release and explicitly provides local-use instructions. This supports technical availability independent of OpenAI’s service.
- Publisher
- Mistral AI
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Supports the claim: BigScience Workshop / arXiv,
hundreds of individuals
Why it matters here: The report documents distributed participation in the project that developed BLOOM, with published training methods. It supports a substantial community of relevant expertise, although participation does not mean every contributor could independently train an entire model.
- Publisher
- BigScience Workshop / arXiv
- Kind of source
- Preprint, not peer reviewed
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, The desperate race to save Generative AI
To begin with, the threat is empty: open source LLMs are already out in the wild, and a large number of people know how to build them.
- Where
- paragraph 9
- Published
- 2024-01-08
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-the-desperate-race-to-save-generative-b5da50-a1: 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-the-desperate-race-to-save-generative-b5da50-a1.