California’s Democratic-controlled Legislature closed its 2026 session on August 31 with a surge of AI regulation heading to Governor Gavin Newsom’s desk. More than 20 bills touching artificial intelligence (covering everything from workplace automation to emotion surveillance to AI-driven layoffs) now await the governor’s signature or veto. He has until September 30 to decide.

For business owners, HR teams, and anyone deploying AI tools with California employees, this is not background noise. These bills carry real compliance obligations, and some of them are substantially broader than anything the state has passed before.

The Bill That Gets the Least Attention but Matters Most

While much of the commentary has focused on SB 947 (the “No Robo Bosses Act”), a quieter bill may carry heavier practical weight for most businesses.

SB 951 revises California’s Worker Adjustment and Retraining Notification requirements, the state’s version of the federal WARN Act, to specifically address AI-driven workforce displacement. Under the bill, if a mass layoff, relocation, or termination is caused in whole or substantial part by an AI system or other automated technology replacing or automating employment positions, employers face new obligations.

The key change: the notice period for AI-driven layoffs would be extended from the standard 60 days to 90 days for events affecting 25 or more workers, or 25 percent of the workforce, whichever is smaller.

That is not just a longer timeline. The bill also requires employers to include specific information in the notice: which job functions are being automated, which AI systems are involved, and what measures the company is taking to support displaced workers. This level of disclosure goes beyond anything currently required under existing state or federal law.

If signed, SB 951 would make California the first state to establish AI-specific layoff notice requirements.

SB 947: Still on the Table

For readers who followed our coverage from August: SB 947, the No Robo Bosses Act, cleared both chambers with stronger-than-expected margins (28-10 in the Senate, 53-14 in the Assembly) and is now waiting on the same September 30 decision.

The bill would bar employers from relying solely on automated decision systems to fire or discipline workers. A human must independently review any AI-assisted termination or disciplinary decision. The bill would take effect July 1, 2027 if signed.

Newsom vetoed a nearly identical predecessor bill, SB 7, in October 2025. Whether the revised language is enough to change his mind is the open question. He is widely seen as a potential 2028 presidential candidate, which makes this decision politically significant beyond California.

Two More Bills to Know

Two other bills in the package are worth flagging:

Emotion surveillance: One bill explicitly bans California employers from using AI tools that monitor or infer employee emotions, mental states, or physiological responses as part of workplace monitoring. This directly affects certain HR technology platforms and productivity monitoring tools that have moved into the emotion-sensing space.

AI in hiring disclosures: A transparency bill would require employers who use AI tools in hiring decisions to disclose that use to applicants. Combined with SB 947’s requirements for termination decisions, the full employment lifecycle, from hiring to firing, would be subject to AI disclosure requirements.

What the September 30 Deadline Actually Means

Governor Newsom signing or vetoing each bill is not a one-time decision. With more than 20 AI bills on his desk, plus a broader slate of privacy and social media legislation, he will be making a series of choices about how aggressively California regulates the AI layer of business operations.

His track record from 2025 suggests he is not reflexively opposed to AI regulation, but he has shown a preference for bills that are targeted and operationally workable over broad mandates that he views as vague or counterproductive.

The bills he vetoes will almost certainly return in revised form in 2027. The bills he signs will become compliance requirements with effective dates typically ranging from 90 days to July 2027.

What Businesses Should Do Before October

If you have California employees (remote workers, office staff, contractors covered by state law) and you use AI tools in any part of workforce management, the next three weeks are a useful window to get ahead of this.

Three practical steps:

1. Map your AI-to-workforce touchpoints. Where does AI currently touch hiring, performance management, scheduling, monitoring, or workforce planning decisions? Which tools, which vendors, which processes?

2. Review your notification processes. If you were to conduct AI-driven workforce reductions affecting California employees, what notice obligations do you currently have, and how would 90 days change your planning cycles?

3. Check your HR software vendors. Several HR technology platforms have been quietly integrating AI features (predictive performance scoring, automated discipline recommendation systems, engagement monitoring) that may put you in scope for these bills whether or not you intentionally “deployed AI” in those functions.

What This Means for Business

The wave of California AI bills reflects something broader. Regulators at the state and federal level are no longer waiting for the industry to self-regulate on workforce impacts. They are moving to require disclosure, human oversight, and longer notice windows specifically because AI-driven workforce decisions have accelerated faster than existing law anticipated.

California legislation reliably sets precedent. What passes in Sacramento tends to spread. Organizations that build compliant AI workflows now are in a fundamentally better position than those who treat each new state bill as a one-off scramble.

The most important design principle for AI-assisted operations is not compliance minimalism. It is making sure AI informs human decisions rather than replacing them. That is what the No Robo Bosses Act is trying to enforce legally. It is also just good practice: AI that supports human judgment rather than bypassing it produces better outcomes and creates far less organizational and legal risk.

If you are building AI-powered operations and want to design them in a way that holds up under regulatory scrutiny, our discovery call is a practical starting point.