Bossware’s Countdown: California’s Workplace AI Rules Go Live in Under 90 Days

On October 8, the UCLA Labor Center and the UC Berkeley Labor Center published a report with a blunt title: Bossware Comes to California. Drawing on existing research and interviews with 25 California workers, the report describes how digital monitoring tools track employees, score performance, and feed automated decisions that workers often cannot see or challenge. Eight days earlier, Governor Gavin Newsom had signed four laws aimed squarely at those practices. The first three take effect on January 1, 2027, which means California employers are now inside a roughly 80-day compliance countdown.

For workers and managers alike, this is the moment when abstract debates about workplace AI become concrete policy. The question is no longer whether bossware-style software is widespread. It is what the rules require, when, and who has to act.

What the January 1 laws actually do

The January 1, 2027 deadline covers three distinct laws, each aimed at a different corner of workplace surveillance and automation.

SB 951 rewrites part of the California Worker Adjustment and Retraining Notification Act (Cal/WARN) for the automation era. When a covered mass layoff, relocation, or termination is caused wholly or substantially by AI or automation replacing positions, the employer must add technology-specific disclosures to the notices sent to employees and the Employment Development Department. The notice must carry the statement "This notice is for a technology displacement" at the top, plus the number, job classification, and work location of affected employees, the job functions being automated, and the category of AI or automated technology involved. Cal/WARN generally requires 60 days of advance notice, and violations can bring civil penalties of up to $500 per day. The Employment Development Department must also deliver a one-time report to the Legislature by January 1, 2028 on AI’s effects on hiring practices.

AB 1883 bans AI-enabled workplace surveillance tools that infer or predict an employee’s emotional state, or that collect neural data by measuring the central or peripheral nervous system. The definition of covered tools is broad: any system that collects information about employee activities, communications, biometrics, or behavior without direct human observation, including video and audio surveillance, time tracking, geolocation, and photo-optical systems. The law reaches private employers, government bodies, the University of California, and California State University, making it broader than the state’s privacy-law rules on automated decisionmaking. Limited exceptions exist for specified safety, aircraft, national security, military, space, and defense purposes. The Labor Commissioner and public prosecutors can enforce it with penalties of up to $500 per violation, and violations can create exposure under the state’s Private Attorneys General Act.

AB 1331 prohibits workplace surveillance in bathrooms, with specified exceptions, and lets workers leave certain surveillance devices behind when entering one.

When software helps decide who keeps a job, the question is no longer what the technology can see, but what the employer can prove about how the decision was made.

The one that waits until July

A fourth law, SB 947, the No Robo Bosses Act, takes effect July 1, 2027. It prohibits employers from relying solely on automated decision systems when disciplining or firing a current employee. If an employer primarily relies on such a system’s output, a human must corroborate the decision using the data behind it or other supporting evidence, and the employee must receive a plain-language notice explaining that an automated system was used, that a human reviewed it, and how to contact a person for more information. Employers may not use these systems to infer a protected characteristic under anti-discrimination law or to punish workers for exercising legal rights. In any dispute, the employer bears the burden of showing compliance. It is not a blanket ban on workplace AI, only on systems making discipline or termination calls without a human in the loop.

What the researchers found

The Bossware Comes to California report offers a ground-level view of the technologies the laws target. The tools range from familiar to futuristic: a phone app tracking a delivery driver’s route, software measuring how quickly warehouse workers finish tasks, activity logging on office computers, automated rankings that flag workers who fall behind targets, and AI products that claim to infer a worker’s emotional state from behavior.

The researchers note that these tools do not all do the same thing. Tracking a delivery route is different from using a computer-generated score to help decide whether someone should be disciplined. The core finding is about uncertainty: workers often do not know what information their employers collect, how long it is kept, or how it feeds into decisions about scheduling, performance reviews, or termination. The authors caution the interviews document experiences, not prevalence. Still, the pattern is the gap the new laws try to close: opaque systems making consequential calls.

The 80-day sprint for employers

Law firms advising California employers are already publishing countdown checklists, and the advice converges on the same three steps. First, inventory every employee-monitoring, productivity, wellness, engagement, wearable, and surveillance tool in use in California. Second, ask vendors directly whether any tool measures neural data or infers emotion, mood, stress, fatigue, engagement, or attentiveness, and document any claimed statutory exception. Prohibited functions must be disabled, modified, or discontinued before January 1, 2027. Third, update Cal/WARN notices and build an internal process for deciding whether a planned reduction is caused wholly or substantially by AI or automation.

The discipline and termination rules get a longer runway to July 1, 2027, but the work starts now: identify every automated score, dashboard, attendance metric, and analytics output used in discipline decisions, and design a human-review step that documents the corroborating evidence and the reviewer’s identity. Employers covered by both these laws and the California Consumer Privacy Act’s automated decisionmaking rules must satisfy each regime separately.

What workers can do before the deadlines

The new protections do not require workers to wait. Employee handbooks and notices about work-issued devices often already describe what is collected, so reading them closely is the first step. If an automated score or performance assessment seems wrong, keeping personal work records and asking a manager or human resources department how the decision was made is now backed by a legal framework: starting July 1, 2027, employers must hand over a meaningful description of the data an automated system used, on request. For help understanding the protections, the California Labor Commissioner’s Office publishes information about labor protections and complaint processes.

California is not alone

The national picture is moving in the same direction. In New Jersey, a bill introduced in May 2026 would bar employers from using AI-based monitoring of a worker’s residence, personal vehicle, or leased property, ban monitoring in bathrooms and locker rooms, forbid compelling workers to install monitoring software on personal devices, and prohibit using AI to make hiring or firing decisions. Its Senate version cleared the Senate Labor Committee on October 1, 2026 in a 5-0 vote. Labor groups support it; business lobbies warn of liability for routine technology use.

Legal analysts note that California’s No Robo Bosses Act is narrower than comparable frameworks in Colorado and the EU, which regulate a wider set of employment decisions.

For now, California sets the pace, and the calendar is doing the talking.