On September 30, 2026, California Governor Newsom signed SB 947, prohibiting employers from firing or disciplining employees based solely on AI algorithms; human review corroboration is required, and laid-off employees have the right to be informed and to a contact person. The law takes effect in July 2027.
Factual Reconstruction
The full name of SB 947 is the “No Robo Bosses Act.” It explicitly prohibits companies from using “automated decision systems” as the sole basis for termination or disciplinary action, and also limits AI’s participation as a “primary tool” in such decisions. If employers rely “primarily” on AI output, a human reviewer must corroborate it with additional information such as management assessments, peer feedback, and personnel files. Affected employees must receive written notice explaining that AI is “primarily used” in the decision, a description of the data used by the system, and a human contact who can provide further explanation.
The bill was introduced by California State Senator Jerry McNerney. In 2025, it won overwhelming support in both chambers of the state legislature but was vetoed by Newsom. When it was reintroduced in 2026, the advance notice requirement was removed and protections for gig workers were stripped out. Newsom had previously signed a broader executive order to address potential risks of AI models and establish an independent evaluation framework.
Mechanism Breakdown
The core of the bill is elevating “human-in-the-loop” from an industry consensus to a legally mandatory obligation. In the past, AI systems could independently generate termination or disciplinary recommendations; now a human reviewer must cross-validate them using non-AI information. Employees gain the right to know, including an explanation of AI use, data sources, and a human contact, which directly changes the weight of automated output in HR processes.
The bill does not cover all AI tools; it targets only termination or disciplinary decisions that rely “primarily” on AI. Companies need to adjust existing automated management software to ensure human intervention points remain in the decision chain while establishing notification and recordkeeping mechanisms.
Industry Impact
As California is the largest labor market in the United States, this move will directly affect the HR systems of technology, retail, and logistics companies. Meta faces a lawsuit from former employees alleging that an AI-assisted system was used for layoff screening, and Walmart and Amazon employees have also expressed concerns about automated HR decisions. After the law takes effect, companies will need to invest resources in overhauling decision-making processes, potentially increasing compliance costs.
On the labor union side, California Labor Federation President Lorena Gonzalez said that worker organizing drove this result and changed the national narrative on AI regulation. Corporate lobbying had previously stalled the 2025 version; after modifications, pressure has eased somewhat, but companies still need to respond to new compliance requirements.
Strategic Assessment (Analysis, Not Fact)
From a labor law perspective, this bill draws a legal red line for AI automated decision-making moving from “recommendation” to “sole basis.” If companies continue to rely on pure algorithmic models, they will face legal risk and need to embed human review into HR systems, which may accelerate the spread of hybrid human-machine decision-making architectures.
For multistate employers, California’s rule may become a template for other states, prompting national HR vendors to adjust product features in advance. In the long run, the bill strengthens the positioning of AI as a tool rather than a decision-maker, but the actual enforcement effect depends on regulatory details after the law takes effect in July 2027 and the speed of corporate adaptation.
© 2026 Winzheng.com 赢政天下 | 转载请注明来源并附原文链接