Anthropic CEO Dario Amodei says company is not advocating for a ban on open-weight AI models

Anthropic CEO Dario Amodei told CNBC on July 27 that the company is not advocating for a ban on open-weight AI models, pushing back against a narrative that has surrounded the safety-focused AI developer. The clarification comes as the debate over open versus closed AI models intensifies globally.

The statement is significant because Anthropic has long positioned itself as one of the strongest advocates for AI safety. The company has repeatedly called for rigorous safety evaluations of frontier AI models and has warned about catastrophic risks from advanced AI systems, leading some observers to assume it would support restrictions on open-weight distribution.

Proponents of open-weight models argue that unrestricted access is essential for academic research, transparency, and reproducibility in AI safety research. Anthropic's previous safety advocacy was interpreted by some as implicitly opposing the open-weight approach.

Amodei's latest remarks clarify the distinction: Anthropic is concerned about safety risks from specific high-capability frontier models, not opposing open-source principles broadly. This distinction is critical as regulators worldwide grapple with how to approach open-source AI in emerging policy frameworks.

The statement may also help ease tensions between Anthropic and the open-source AI community. The company's product strategy — including the release of Claude model weights to select partners — has at times been viewed as falling on the closed side of the spectrum. Amodei's comments suggest a more nuanced position that acknowledges the legitimacy of open-weight development.

Looking ahead, the key question is whether Anthropic will eventually release its own open-weight models or contribute more directly to the open-source AI ecosystem. The company's product decisions will carry more weight than any single public statement in defining its true stance on openness.