Summary
After nearly seven months of lawmaking, California legislators ended this session without the passage of a bill regulating the development or deployment of artificial intelligence (AI) systems. This comes as a surprise to many as California has historically led the Nation on consumer protection issues. For example, California passed the first in the nation omnibus consumer privacy law, the California Consumer Privacy Act, and has led the charge on children’s online privacy with the passage of the Age Appropriate Design Code Act. However, both of those laws have been challenged in some manner in the courts, and the decision not to pursue AI legislation may reflect a reduced appetite for trailblazing laws that inevitably draw challenges, as well as the administrative burdens of drafting regulations, standing up enforcement agencies and finding the resources to actually engage in enforcement.
On September 6, however, California Governor Gavin Newsom signed an
Executive Order on government procurement of generative AI (“GenAI”). The Order requires that various State agencies and departments examine the “most significant, potentially beneficial use cases for deployment of GenAI tools by the State” as well as the “potential risks to individuals, communities, and government and state government workers, with a focus on high-risk use cases, such as where GenAI is used to make a consequential decision affecting access to essential goods and services.” Agencies must report their findings within 60 days of issuance of the Order. The Order also requires that the State issue “general guidelines for public sector procurement, uses, and required training for use of GenAI,” and consider pilots of GenAI projects to be tested in approved environments, or “sandboxes.” Viewed holistically, the Order signals that California, which by some accounts is home to 35 of the world's top 50 Al companies, is going to take a measured and “pro innovation” approach to regulating GenAI.
The lack of movement by the California legislature has implications well beyond the confines of California. The failure to advance a broad AI bill has further cleared the path for Congress to act without fear of obstruction from the California Congressional delegation, as we saw with the American Data Privacy and Protection Act - the first real attempt at a national omnibus consumer privacy law - which was scuttled over California’s preemption demands.
To that end, Congress held competing hearings on AI over the course of this week. On Tuesday, a Senate Commerce subcommittee held a hearing on the "Need for Transparency in Artificial Intelligence.” Senate Majority Leader Chuck Schumer, D-N.Y. separately hosted highly publicized closed-door "AI Insight Forums", with appearances from Meta CEO Mark Zuckerberg and Tesla CEO Elon Musk, among other tech leaders. And finally, a House Oversight subcommittee held a hearing on Thursday examining the potential risks in federal agency adoption of AI and the adequacy of safeguards to protect individual privacy and mitigate bias. Members of Congress are working to close the significant technological knowledge gap, but it remains to be seen how soon this learning will translate to lawmaking.
In the meantime, state legislatures continue to advance bills that regulate the development and use of AI by private industry. Many of these bills are omnibus consumer privacy laws that include AI-related provisions within a broader privacy framework. States like New York, New Jersey, and the District of Columbia have followed in the footsteps of New York City and have also advanced bills that target the use of AI in the recruitment and employment contexts, which has become an area of focus for state and federal regulators.
For up-to-date information on AI legislation at the state level, please see our 2023 state-by-state AI legislation snapshot.
[View source.]
