California bans AI tools that track employee emotions or neural data

New limits on workplace AI

California is expanding its oversight of artificial intelligence in the workplace with the passage of several new laws. These regulations aim to place specific limits on how companies use automated systems to manage and monitor their staff. The state is moving toward a framework where AI compliance becomes a primary concern for human resources departments.

A major component of this shift is AB 1883, which specifically targets the use of AI-powered surveillance. This law restricts how employers can use technology to gather data on their workers' internal states. It marks a transition from general data collection to a more specific protection of psychological and biological privacy.

While these rules are currently specific to California, experts suggest that other states may follow this lead. The move signals a growing interest in regulating the intersection of high-tech monitoring and labor rights. Employers will need to audit their existing software to ensure they meet these upcoming standards.

What is banned under AB 1883

The new law prohibits employers from using surveillance tools that use AI to recognize or make inferences about an employee's emotional state. This means companies cannot use software to guess how a worker is feeling or predict their mood based on their behavior. The goal is to prevent invasive psychological profiling in the office or on the job site.

Additionally, the law bans the collection of neural data in the workplace. Neural data is defined as information generated by measuring the activity of an employee's central or peripheral nervous system. This is distinct from information that is simply inferred from non-neural sources like speech or movement.

The legislation defines a workplace surveillance tool broadly to ensure companies cannot find loopholes. It includes any system, application, or device that collects employee data, communications, or biometrics. This covers everything from video and audio surveillance to geolocation and electromagnetic tracking systems.

Exceptions for workplace safety

The law does include a specific exception for monitoring that is necessary to ensure safety. This means that employers are not prohibited from using surveillance tools if the primary goal is to keep workers out of harm's way. This distinction is important for industries that involve high-risk physical environments.

For example, the law may allow for systems that monitor fatigue in drivers or other safety-sensitive positions. These tools are designed to prevent accidents rather than to judge an employee's emotional state or personality. The focus remains on physical security rather than psychological monitoring.

Despite these exceptions, the broad definitions in the law mean that companies must be careful. They will need to clearly distinguish between safety-related data and data that could be interpreted as emotional tracking. Failure to maintain this distinction could lead to legal complications for the employer.

The broader California AI landscape

AB 1883 is part of a larger suite of AI-related laws recently signed in California. Other regulations, such as SB 947, also impact how companies use automated decision systems. That law, known as the No Robo Bosses Act, prevents employers from relying solely on automation for disciplinary actions or terminations.

Another law, SB 951, addresses how companies handle mass layoffs. It requires employers to provide specific types of notice if a layoff is caused in whole or substantial part by AI. These rules collectively aim to ensure that humans remain in the loop for significant employment decisions.

By layering these laws together, California is creating a complex regulatory environment for AI. Companies must now navigate multiple requirements regarding hiring, firing, and daily monitoring. Each piece of legislation adds a new layer of responsibility for how technology is deployed in the workforce.

Consequences and compliance

Violations of AB 1883 can carry significant financial and legal weight for businesses. The law allows for civil penalties of up to $500 per violation. In addition to these fines, companies may face injunctive relief and punitive damages during enforcement actions.

The Labor Commissioner or a public prosecutor can bring these actions against companies that fail to comply. This means that legal costs and attorneys' fees can also be recovered in enforcement cases. These penalties are designed to discourage the use of invasive AI tools and encourage strict adherence to the new rules.

The overarching message for businesses is to proceed with caution when using AI for employment-related tasks. Employers are encouraged to avoid relying solely on automated systems for making decisions. When AI is used in the workplace for any reason, companies must ensure the specific tools they choose comply with the law.