Introduction
In 2023, a job applicant named Derek Mobley sued Workday for its use of an AI-powered hiring system, claiming that the AI was discriminating against older applicants and those with disabilities. The case brought deeper scrutiny on the use of AI systems in making hiring decisions. This issue ties into a recent bill proposed by California State Senator Jerry McNerney on March 6, 2025, titled the No Robo Boss Bill (S.B. 7), which addresses the increasing use of AI and automated decision systems (ADS) in the workplace. This bill seeks to regulate the use of AI in making critical employment decisions such as hiring, firing, promoting, and disciplining employees. California S.B. 7 is not the first proposed bill to regulate AI in the workplace—an Illinois bill, H.B. 3773, has similar aims. However, the No Robo Boss Bill further regulates the use of AI by prohibiting automated, real-time managerial decisions and requiring human oversight in workplace monitoring systems.
After receiving several amendments, S.B. 7 passed the California Senate on June 2, 2025, and is currently under consideration in the Assembly. One key focus of the bill is on automated decision systems (ADS), which refers to any system, software, or process driven by AI or machine learning that makes, recommends, or aids employment-related decisions. S.B. 7 proposes that when an employee appeals a decision made by an ADS, the employer must respond within fourteen days and assign a human reviewer to review the decision. Furthermore, in response to matters like Mobley v. Workday, S.B. 7 prohibits employers from using personal attributes such as immigration status, religion, or health information in employment decisions. It also bans the use of algorithms to conduct predictive behavior analysis aimed at determining an applicant’s future employment status.
The use of AI in managing administrative tasks in the workplace—often referred to as “bossware”—has sparked growing debate, with over 40 percent of workers reporting experiences with some form of automated oversight. As a result, labor unions such as the California Federation of Labor Unions strongly supports S.B. 7, believing it would make an essential step towards protecting workers’ rights, dignity, and safety. However, AI is also often considered as a tool to boost efficiency, and employer groups worry that S.B. 7 would place additional administrative burdens on small businesses and hinder the pace of innovation.
Arguments in Support of the Bill
Supporters argue that S.B. 7 would strengthen worker protections by requiring human oversight in automated employment decisions, since the bill seeks to prevent the use of AI in decisions regarding hiring, firing, and promotions without human review. For example, companies have been using AI to scan office emails for signs of burnout or dissatisfaction, or to filter job candidates by searching thousands of online sources for potential red flags like references to violence or drug use. These behaviors are a major concern for advocates, who argue that these uses of AI are not only overly intrusive but also susceptible to error and bias. They argue that human oversight is necessary to prevent errors, discrimination, and invasive practices, as well as to ensure that employment and termination decisions are made with transparency.
Another argument in support of S.B. 7 is related to concerns over privacy and ethical issues. The policy analysis published by the Assembly Committee indicates that, while 40 percent of workers in the United States experience some form of automated task management, Black and Latino workers are subject to these systems at disproportionate rates—63 percent and 52 percent—compared to only 35 percent of white workers. This data demonstrates the potential for automated systems to discriminate against certain racial and ethnic groups.
Moreover, these systems often make inferences about workers based on sensitive data, raising significant privacy concerns. S.B. 7 seeks to prevent AI from being used to infer a worker’s immigration status, health history, or emotional state, directly addressing key privacy concerns. With stronger limits on data use and predictive analytics, S.B. 7 seeks to reduce such inequities and protect workers’ rights to privacy. Lorena Gonzalez, the president of the California Federation of Labor Unions, stated, “We have to set up guardrails against every kind of surveillance and AI tool to ensure that workers have the privacy and respect and autonomy that they deserve.”
Lastly, proponents of the bill say it reinforces workplace accountability by guaranteeing that the ultimate authority to review any employment-related decisions rests with humans, not algorithms. The bill mentions that decisions can be based on sources such as “supervisory or managerial evaluations, personnel files, work product of workers, and peer reviews,” which are human-conducted analyses that authentically reflect an employee’s performance. By prioritizing these human-centered decision-making processes, S.B. 7 aims to preserve accountability and ensure that AI is not the core authority in these employment decisions.
Arguments Against
S.B. 7 has also raised some concerns. Firstly, opponents argue that the bill would impose operational burdens on companies that slow down everyday human resources tasks such as hiring and performance reviews. It’s estimated that if the bill passes, California employers may need to hire between 2,150 and 4,300 new full-time HR professionals to manage appeal processes and data access requests, which could cost businesses between $523 million and $1 billion annually. Since the bill would require employers to send a detailed, personalized notice to every applicant rejected by an automated system, companies could face the task of handling thousands of subsequent appeals from those applicants on a daily basis. While this requirement may increase transparency for applicants and help them feel acknowledged in the hiring process, critics argue it could significantly undermine operational efficiency and even reduce the timeliness of hiring decisions.
Secondly, critics argue that strict scrutiny of AI in the workplace may limit the potential for valuable innovations in staff organization and management. They warn that broad restrictions such as those in S.B. 7 may “end up broadly sweeping up common uses of AI that are actually intended to help employees, including employee retention, employee satisfaction, and other similar goals.” Moreover, critics highlight that banning AI does not guarantee a full prohibition of algorithms, as simpler mathematical formulations may still be used in hiring decisions. Such basic formulations may not be able to achieve the ideal outcomes such as reducing hiring costs and efficiently matching applicants to roles. As a result, the bill could inadvertently sacrifice both efficiency and fairness in pursuit of regulation.
Third, critics raise concerns about S.B. 7’s vagueness and challenges around its implementation. They argue that the bill lacks clear definitions of what constitutes adequate “human oversight,” which could be applied inconsistently and rely heavily on individual interpretation. This uncertainty is problematic for both AI providers and employers, who may face conflicting standards on how to comply fully with the policy. For example, even though the bill grants rejected applicants the right to appeal decisions, employers may be unable to reverse those decisions if the position has already been filled. This scenario raises difficult questions about balancing the rights of all applicants, which may lead to unintended fairness issues for both newly hired employees and those seeking to appeal.
Conclusion
Today, AI regulation in the workplace is expanding to address related areas such as data protections, transparency, and employee rights. While S.B. 7’s main focus is on ensuring human oversight in supervising and reviewing employers’ operational usage of ADS, it reflects a broader movement toward regulating how algorithms shape workplace dynamics. The policy conversation is expected to evolve with technology, with further legislative efforts anticipated to be centered on algorithmic fairness, bias auditing, and worker data protections as these technologies become more embedded in workplace management.
In a statement to the Members of the California State Senate, Governor Gavin Newsom vetoed S.B. 7 on Monday, October 13th. Newsom stated that while he “share[d] the author’s concern” about the potential harm by unregulated use of Automated Decision Systems (ADS) unto workers, he disagreed with S. B. 7’s “unfocused notification requirements” for businesses. Furthermore, his veto is also based on the bill’s “overly broad restrictions” for employers’ use of ADS tools.