California has become the first state in the US to pass a law that directly limits how employers can use artificial intelligence (AI) in firing or disciplining workers. signed SB 947, known as the ‘No Robo Bosses Act’, last week, setting new rules that will take effect on1 July, 2027.
The law makes it clear that companies cannot lean entirely on algorithms to decide whether someone keeps their job. Automated-Decision Systems (ADS), including AI tools, can no longer be the sole or main factor in disciplinary or termination decisions. Employers must disclose when AI is used, explain what personal data was fed into the system, and—if AI plays a major role in a firing—provide a human reviewer to check the decision. That reviewer must back up the decision with real evidence such as performance evaluations, disciplinary records, or witness testimony.
This move comes as industries wrestle with the growing role of AI in workplace management. While federal law already covers AI under existing anti-discrimination protections, and other states have banned AI-driven bias in hiring and benefits, California’s new law goes further. It directly restricts AI’s authority in discipline and termination, ensuring that human judgement remains central when livelihoods are at stake.
The bill applies to all employers in California, regardless of size, though gig workers are excluded. Newsom had vetoed an earlier version in 2025, citing overly broad notification requirements, but the revised text narrowed those provisions before winning approval.
If companies violate the law or retaliate against employees who assert their rights, they could face civil penalties from the California Labour Commissioner’s Office. By drawing a line between human oversight and machine output, SB 947 sets a national precedent and signals that while AI may assist in management, it cannot replace human accountability in decisions that affect people’s jobs.