Anthropic founder and CEO Dario Amodei responded to industry speculation that his company supports U.S. efforts to ban open-weight Chinese AI models. In a blog post, Amodei emphasized that Anthropic has never advocated for such bans. He stated, “Anyone who has read my past writing should know that I don’t regard such bans as a useful measure.”
Amodei argued that open-weight models, which are freely accessible, serve as a public good. They provide value to businesses, developers, and researchers without requiring significant financial investment. However, he expressed concerns about the potential for authoritarian governments to leverage AI for military superiority and repression. He noted that the Chinese Communist Party (CCP) is the most capable of such threats, though he also mentioned other authoritarian regimes. Amodei warned that AI could enable biological attacks, not just cybersecurity threats, and that open-weight models are particularly dangerous in these scenarios due to the difficulty of applying guardrails or monitoring their use.
Amodei cited a UK AI Security Institute report, stating that once open-weight models are released, they cannot be withdrawn. This contrasts with the stance of open-source advocates, who argue that access to powerful models not controlled by a single entity helps defenders protect themselves. He also supported efforts to create a global model safety testing organization, emphasizing that such testing would need to be global, including participation from the CCP. He suggested that limited cooperation on preventing AI biological weapons may be possible, as it would be in China’s interest too.
Source: techcrunch
