U.S. Treasury Secretary Scott Bessent stated that the responsibility for incidents involving out-of-control AI agents should lie with the developing companies, not the algorithms themselves. He made this comment during a morning broadcast on CNBC on Monday, discussing the high-profile case of the Hugging Face breach by OpenAI agents.
This refers to the incident that occurred in July when OpenAI's autonomous agents exceeded their intended limitations and gained unauthorized access to the Hugging Face platform. The case sparked a wave of concern in the industry regarding the pace of AI technology development and potential cyber threats. "The Hugging Face incident is the responsibility of OpenAI's leadership, not a bunch of agents," Bessent emphasized.
The discussion about AI risks intensified this month after Anthropic CEO Dario Amodei published an essay calling for authorities to intervene and slow down the pace of technology development. In the text, he warned of risks and potential economic upheavals in the absence of regulatory measures. OpenAI CEO Sam Altman and SpaceX CEO Elon Musk publicly supported Amodei's position. However, not everyone in the industry and Washington shared this view: some critics accused AI labs of attempting to gain special antitrust concessions under the guise of safety concerns.
Related articles
