Anthropic CEO Dario Amodei has publicly clarified his company's position, asserting that it has "never advocated for a ban on open-weights models." This statement, delivered via a blog post on Monday, aims to counter growing criticism across the tech industry suggesting the AI lab seeks undue control over artificial intelligence's future trajectory.
Amodei's clarification follows the release of a significant letter late last week by a coalition of major tech players, including Nvidia, Microsoft, Meta, and Palantir. This letter urged policymakers to steer clear of "premature restrictions" on open-weight models, which are AI models that users can freely download, modify, and operate on their own infrastructure. The debate is particularly timely as Chinese startups are dominating this market, prompting some government officials to consider potential bans or restrictions within the U.S.
Dario Amodei, chief executive officer of Anthropic, at the AI Impact Summit in New Delhi, India, on Thursday, Feb. 19, 2026. Ruhani Kaur | Bloomberg | Getty Images
Anthropic is renowned for its proprietary Claude model family, which it licenses to businesses. Its primary competitor, OpenAI, which also predominantly develops closed models, notably signed the open-weight letter post-publication. Anthropic, however, did not. Amodei explicitly stated, "To summarize my and Anthropic's position, we have not and are not advocating for a ban on open-weights models as a category." He stressed that efforts should instead concentrate on preventing powerful chips from falling into authoritarian hands, curbing industrial-scale distillation, and mandating safety testing for all sufficiently capable models, whether open or closed.
Distillation, an AI training technique where a smaller, less advanced model is built using outputs from a larger, existing model, has been a contentious topic. Last month, Anthropic accused China's Alibaba, developers of the Qwen model series, of conducting what it termed "the largest known distillation attack" against its models in a letter to the U.S. Senate Committee on Banking, Housing, and Urban Affairs.
The tech coalition's letter on Friday suggested that concerns regarding unlawful distillation should be handled via "targeted legal and commercial frameworks," a point on which Amodei concurred. He also aligned with other aspects of the letter, acknowledging that open-weight models empower customers with greater control, broaden access to the AI economy, and enhance competition in certain applications. However, he diverged on the notions that open-weight models inherently favor cyber defenders over attackers or simplify the development of safeguards for users.
Despite these nuanced disagreements, Amodei reiterated his clear stance against banning open-weight models, stating, "Protectionist bans would not address my most serious national security concerns."
WATCH: Anthropic’s new AI model rivals Fable 5 and is cheaper as businesses fret about costs
/https://media.cnbc.com/dims/origin/108343257/Spider-Man-Brand-New-Day.jpg "Spider-Man Smashes Box Office Records with ‘Brand New Day’ Preview Haul")
