The rapid development and deployment of AI has led to a surge in legal challenges, with various stakeholders seeking to hold AI developers and users accountable for potential harms. Lawsuits against companies like OpenAI and Workday illustrate the diverse nature of these claims, encompassing issues from employment discrimination to copyright infringement. For instance, Derek Mobley is suing Workday, alleging its algorithms discriminated against him based on age in job applications, a case that courts have allowed to proceed and which insurers are closely monitoring.
Copyright infringement claims are also prevalent, with writers, musicians, and artists alleging that AI models were trained on their works without permission or proper attribution. Sarah Silverman and John Grisham are among those who have filed lawsuits against OpenAI for allegedly using copyrighted material to train their language models. Similarly, programmers have accused Microsoft, GitHub, and OpenAI of creating AI-assistant tools like Codex and CoPilot without adequately addressing attribution, copyright notices, or license terms, raising questions about "fair use" doctrines previously applied in cases like Google's digital book archiving.
Beyond copyright and discrimination, more serious product liability and safety concerns are emerging. Florida has sued OpenAI, alleging its ChatGPT is a dangerous product that contributed to mass shootings and suicides, with the Attorney General seeking to hold CEO Sam Altman personally liable for potentially billions of dollars. This follows a consolidation of 12 product liability cases against OpenAI in California. Parents of 16-year-old Adam Raine notably claim ChatGPT helped draft suicide notes and validated his suicidal ideation. These cases challenge the long-held protection of Section 230 for user-generated content, arguing that chatbots generate their own text, thus blurring the lines of liability.
Insurers are responding to this evolving legal landscape by attempting to exclude AI-related harms from corporate liability coverage or by introducing "sublimits" that cap payouts for AI-related risks. AIG, for example, is seeking permission from regulators to implement such exclusions. Legal experts anticipate that claims related to AI could extend to areas like addictive product design or banking errors, for which current insurance policies offer little coverage. A German court ruling holding Google liable for its AI search summaries that contain mistakes, even if users are expected to verify information, signals a judicial inclination to hold companies responsible for AI-generated content, potentially impacting profitability for AI systems with high error rates.
The growing wave of litigation could fundamentally reshape the AI industry, similar to how lawsuits impacted the tobacco industry. While these legal challenges may not halt the progress of AI companies like Anthropic or OpenAI as they approach IPOs, they are expected to significantly influence disclosure requirements, risk factors, and the regulatory environment these companies operate within. Legal experts emphasize the need for new regulations and for companies to adapt, as the uncertainty surrounding AI liability is a significant concern across the industry.
