Governor Gavin Newsom signed SB 947, the No Robo Bosses Act of 2026, on Sept. 30, making California the first state to bar employers from relying solely on automated systems to fire or discipline workers. It was one of more than a dozen bills he cleared the same day as the state’s Sept. 30 deadline rundown closed out.

The law takes effect July 1, 2027. Per CNBC, it requires a human reviewer to corroborate decisions where AI is “primarily” relied on, and workers must receive written notice that AI drove the call, a description of the data considered, and a human contact. Enforcement runs through the California labor commissioner, state attorney general, or local prosecutors. Companion measures signed the same day ban emotion-prediction from biometric data and require written notice when AI drives mass layoffs.

The path here was narrow. Newsom vetoed an earlier version in 2025; author Sen. Jerry McNerney stripped pre-notification requirements and gig-worker coverage to secure signature this cycle. The Chamber of Progress has already flagged the central drafting gap, warning that “primarily relies” is undefined, which will shape every enforcement fight to come. It’s the familiar arc of California tech regulation: landmark on paper, litigated into shape over years. The same dynamic defined the state’s independent AI auditor regime and this week’s sensitive-data veto.

For small-business owners, the scope matters more than the headline. SB 947 attaches to employment decisions. AI used to identify prospects, draft outreach, or produce marketing content isn’t an employment decision and doesn’t trigger the law’s human-review, notice, or data-summary obligations. See the full bills-still-alive tracker for what cleared and what didn’t.

LemonLime, which prepares sales and marketing work for small businesses and requires customer approval on every send, operates outside the automated-decision-system scope the new California laws target.

Sources

Sources