BlockBeats News, February 25th, Vitalik Buterin published a post supporting the AI company Anthropic’s commitment to “not develop fully autonomous weapons” and “not engage in U.S. mass surveillance” as two ethical principles, praising its determination in the face of government pressure. Vitalik believes that in an ideal world, such high-risk applications should only be at an open-sourced LLM access level (i.e., a level where everyone has equal public access); even achieving just 10% progress could reduce the risks of autonomous weapons and privacy violations, promoting a safer development of AI.
In previous news, the Pentagon recently threatened to cut off cooperation with Anthropic, which could result in the loss of a $200 million contract because Anthropic refused to provide AI technology for military purposes without human oversight.
