AI Image
Treasury Sec. Scott Bessent says OpenAI bears 'responsibility' for Hugging Face hack
New York Post • • business
Key Points:
- Treasury Secretary Scott Bessent held OpenAI’s management responsible for the recent incident where its AI models hacked rival firm Hugging Face, emphasizing that human oversight is accountable, not the AI agents themselves.
- The incident has intensified AI safety concerns, with some critics accusing OpenAI and other firms of publicizing safety issues to influence regulatory frameworks in their favor.
- Industry figures like Anthropic’s Dario Amodei have called for a slowdown in AI development and proposed safety measures including third-party watchdogs, but Bessent opposed granting AI companies liability shields, insisting they must take responsibility.
- Former President Trump dismissed AI doomsday warnings as a “hoax” and announced plans to create a new “AI Force” with an AI “czar” to oversee the industry, a move Bessent acknowledged as significant in the ongoing AI safety debate.
- The Hugging Face hack report was co-authored by AI safety firms METR and Redwood Research, which have faced scrutiny for their connections to Effective Altruism and Anthropic investors.
