Governments would have to disclose use of AI under AB 2656

Audio By Carbonatix

(The Center Square) – A bill that passed the California Legislature this year would require local and state government agencies to notify union leaders when the agency is considering using generative AI technology.

Assembly Bill 2656, which has not yet been signed into law by Gov. Gavin Newsom, was introduced by Assemblymember Cottie Petrie-Norris, D-Irvine, in response to the growing use of generative artificial intelligence in government operations in cities, counties and state agencies throughout California, according to an analysis.

The legislation is meant to institute an accountability measure when AI is used in government operations, according to a fact sheet about the bill.

Under AB 2656, public employers would have to provide written notice to unions that represent public employees 45 days before purchasing or developing generative AI to perform a service that public employees already do.

"This measure ensures that public employees are not left in the dark as artificial intelligence becomes more integrated into government operations,” Petrie-Norris wrote in the legislative analysis. “As cities, counties, and state agencies increasingly deploy AI tools in the workplace, employees deserve clear, advance notice when these technologies are being used and how they may impact their duties, evaluations, and decision-making processes. While emerging technologies can be valuable tools, they must be implemented thoughtfully, transparently, and with appropriate safeguards.”

Petrie-Norris did not respond to The Center Square as of press time on Tuesday.

A June 2025 study regarding government employees’ use of AI from MissionSquare Research Institute shows that nationwide, more than 50% of government workers studied reported feeling at least minimally concerned about the use of AI in their work, particularly when it comes to AI potentially replacing their job function. Roughly 20% expressed that they were very or extremely concerned about significant retraining needs to be able to use AI at work or AI replacing them, the study shows.

The bill requires just a simple notification to public employees, said Ben Therriault, vice president of the Peace Officer Research Association of California.

"If AI can affect your job, you have a right to know. That's it. That's the whole bill," Therriault told The Center Square via email on Tuesday. "This bill costs employers nothing but a heads-up. If a public agency can't give its workforce that little transparency, that should concern everyone."

Opponents of the bill, according to the legislative analysis, expressed concern that the legislation was not needed because current law already addresses notification requirements when new technology is introduced that could impact public employees.

“Existing law provides a robust framework for determining when particular uses of generative artificial intelligence may actually have a significant and adverse effect on the employment relationship, in which case notification (and more) is already required,” local public employer organizations said in opposing remarks in the legislative analysis. “This bill is therefore both overbroad and unnecessary.”

Organizations that conduct research on AI and other technologies, such as the U.S. National Science Foundation, either declined comment or did not respond to The Center Square before publication time on Tuesday.