Former OpenAI Researchers Push for California AI Safety Bill, Challenging Company's Stance
Former OpenAI Researchers Criticize Company’s Opposition to AI Safety Bill
In a surprising turn of events, two former OpenAI researchers have publicly criticized the company’s stance on a proposed California AI safety bill. William Saunders and Daniel Kokotajlo, who previously worked at the artificial intelligence research laboratory, expressed their disappointment in OpenAI’s opposition to SB 1047, a bill that would impose strict safety protocols on AI development, including a “kill switch” mechanism.
The researchers, who resigned from OpenAI due to concerns about the company’s commitment to safe AI development, penned a letter to California Governor Gavin Newsom and other lawmakers. In their missive, they argued that rapid AI advancement without proper regulation risks “catastrophic harm to the public.”
This criticism comes in stark contrast to OpenAI CEO Sam Altman’s public statements supporting AI regulation, including his recent congressional testimony. Saunders and Kokotajlo contend that while Altman voices support for regulation in principle, the company actively opposes concrete regulatory proposals.
OpenAI, however, disputes this characterization. A company spokesperson defended their position, citing a letter from Chief Strategy Officer Jason Kwon to Senator Scott Wiener. In this communication, Kwon explained that OpenAI believes AI regulation should be implemented at the federal level for national security reasons.
The debate between state and federal regulation has become a focal point of the discussion. Kwon argues that a federally-driven set of AI policies would be more effective than a patchwork of state laws. However, Saunders and Kokotajlo counter that waiting for federal action is not viable, pointing to Congress’s apparent reluctance to pass meaningful AI regulation.
The former researchers are urging the California Legislature and Governor Newsom to pass SB 1047, hoping that appropriate regulation will help OpenAI fulfill its mission of building artificial general intelligence (AGI) safely.
As of now, representatives for Senator Wiener and Governor Newsom have not responded to requests for comment on this developing story. The outcome of this debate could have far-reaching implications for the future of AI regulation in California and potentially across the United States.
