Governor Gavin Newsom has issued an executive order to establish a framework for artificial intelligence safety in California. This move initiates a committee process to study AI risks and develop recommendations before the governor's term concludes.
The shift from ChatGPT's 2022 debut to existential risk
The public perception of artificial intelligence has undergone a rapid transformation since the general release of ChatGPT in late 2022. Initially, the conversation centered on the immediate disruption of schools, heatlhcare, and the workforce. However, as reported by the KPBS Roundtable, the dialogue has shifted toward more urgent concerns regarding safety, oversight, and the potential existential risks these systems pose to humanity.
This evolution in sentiment has created a political opening for California to act. By framing AI as a safety issue rather than just an economic one, Governor Gavin Newsom is tapping into a growing demand for advanced AI developers to slow their pace or accept stringent limits. This shift reflects a broader global trend where the initial awe of generative AI is being replaced by a cautious desire for institutional guardrails.
Adopting Illinois' model for independent AI evaluations
Rather than implementing a sweeping law immediately, Governor Gavin Newsom is using this executive order to launch a study and recommendation phase. according to the KPBS Roundtable , the governor is signaling a desire for California to align with the tougher standards seen in Illinois. illinois is currently regarded as a leader in requiring independent, third-party evaluations of AI systems to ensure safety .
The goal of this approach is to eliminate the conflict of interest inherent in current industry practices.. By moving toward the Illinois model, California aims to prevent AI companies from simply assessing their own high-risk models without external scrutiny. Steven Counsel of Business Insider described the order as the beginning of a committee process designed to build a foundation for future, more concrete regulatory actions.
The possibility of a special legislative session in Sacramento
While the executive order provides an immediate path forward, Governor Gavin Newsom is also weighing more aggressive legislative options. Chase de Feliciano Antonio of Politico noted that the governor has considered calling a special session of the California Legislature to address AI safety, even though lawmakers have already adjourned for the year.
A special session in Sacramento would allow for the creation of durable laws, but it carries significant political risk. Lawmakers may disagree on the extent of the restrictions, and any proposal could be rewritten or delayed during the political process. Nevertheless, by initiating this process now, Governor Gavin Newsom is attempting to force AI safety onto the national agenda, effectively challenging the federal government to provide a more direct response to these technological risks.
Who will staff the new California oversight group?
Despite the announcement, several critical details remain unresolved. The KPBS Roundtable highlights that while a new oversight group will be formed, the specific composition of this group—and who will be appointed to lead the recommendations—has not been disclosed. It remains unclear whether the committee will be dominated by state bureaucrats, academic experts, or representatives from the technology industry itself.
Furthermore, the report indicates that the working group is expected to offer recommendations within a few months, but it does not specify the exact criteria that will define a "high-risk" model. Without a clear definition of what constitutes a dangerous AI system, the transition from a study committee to enforceable regulation remains a significant hurdle for the administration in Sacramento.
Comments 0