The California State Assembly has introduced a new bill to regulate artificial intelligence systems. The proposal, A.B. 331, requires developers of AI systems to mitigate inherent bias within automated decision-making tools (ADTs).
The bill seeks to regulate ADTs, which are used to make decisions on any type of eligibility for a benefit or penalty.
While these systems have been traditionally used for credit decisions, their usage has expanded to employment screening, insurance eligibility, and health care decisions. However, ADTs are often trained on biased data, resulting in harm to marginalized communities.
A.B. 331 requires developers and users of ADTs to conduct and record an impact assessment that includes the intended use, the makeup of the data, and the rigor of the statistical analysis. The data reported must also include an analysis of potential adverse impact on the basis of race, color, ethnicity, sex, religion, age, national origin, or any other classification protected by state law.
The bill has passed the body’s Privacy Committee. The California initiative is similar to the recent New York City law requiring bias testing of AI systems that make employment decisions, as well as a D.C. Council proposal for bias testing AI systems used in decision making for employment, credit and housing.
The California Senate has introduced another bill, S.B. 721, that would establish the California Interagency AI Working Group to deliver a comprehensive report on artificial intelligence to the state legislature.
California law already addresses many AI applications, including autonomous vehicles, social media, bots, fake news, drones, privacy, and job displacement. The state, however, does not yet directly regulate AI.
The working group would report to the legislature its recommendations for a definition of AI and determine which state agencies would develop and oversee artificial intelligence policy and implementation. The report would also identify best practices for the use of AI, including privacy and security safeguards, mitigation of bias, and ethical considerations. A sunset provision would end the group’s work on January 1, 2030.
Potomac Legal Group is closely monitoring developments that impact AI and decision making in employment matters. Contact our attorneys should you have questions or concerns about employment decisions or AI and the law.
Leave a Reply