California Committee Discusses AB 316 to Hold AI Developers Accountable for Harm

The Assembly Privacy and Consumer Protection Committee of the California State Assembly convened on May 6, 2025, to discuss Assembly Bill 316 (AB 316), a proposed legislation aimed at clarifying liability in cases involving artificial intelligence (AI). The meeting featured a range of expert testimonies and public comments, highlighting both support and opposition to the bill.

The primary focus of AB 316 is to establish a framework that holds developers and deployers of AI accountable for harms caused by their products, regardless of claims that the AI acted autonomously. Assembly member who introduced the bill emphasized that it does not alter existing liability laws or create new avenues for lawsuits. Instead, it aims to prevent defendants from using the defense that AI independently caused harm, thereby ensuring that those who benefit from AI technology also bear responsibility for its consequences.

Two expert witnesses supported the bill. Ken Wang, a senior policy advocate, argued that as AI technology evolves, it is crucial to clarify that developers cannot evade responsibility by attributing harm to AI. He provided examples, such as incidents involving Tesla's autopilot feature, to illustrate the need for accountability. Ed Howard from the Children's Advocacy Institute echoed these sentiments, warning that the influence of AI, particularly on vulnerable populations like children, necessitates clear accountability measures.

The committee also heard opposition from several stakeholders, including representatives from Technet and Cal Chamber. They expressed concerns that the bill could create legal uncertainties and potentially deter investment in AI technology. Opponents argued that existing tort law adequately addresses accountability and that the proposed legislation might inadvertently impose excessive liability on developers and users, regardless of their actual level of fault or responsibility.

The discussions highlighted the tension between fostering innovation in AI and ensuring public safety and accountability. Supporters of AB 316 maintained that without clear liability, there would be little incentive for developers to prioritize safety in their AI systems. In contrast, opponents cautioned that the bill could lead to unintended consequences, complicating the legal landscape for AI development.

The committee concluded the meeting without a definitive vote on the bill, indicating that further discussions and considerations would be necessary. The outcome of AB 316 remains uncertain as stakeholders continue to weigh the implications of AI accountability in California's evolving technological landscape.