AI and the new frontiers of trade secret protection

A recent federal trade secrets case in the Northern District of California highlights the growing risks of AI in misappropriating proprietary information. The defendant, a former senior employee, used stolen trade secrets to train an AI agent, effectively creating a digital replica of his former boss to compete without years of expertise. The court deemed the 'quarantine folder' used by the defense as insufficient, granting a temporary restraining order and preliminary injunction due to the AI agent's 'existential threat.' Legal experts warn that traditional methods of protecting trade secrets are no longer adequate. AI's ability to scan, ingest, and train on data without user awareness poses significant risks, especially when employees use personal devices. Founders are urged to take immediate action, cutting off system access and consulting legal counsel if they suspect data misuse. Seth Kugler, a partner at Grellas Shah LLP, emphasizes the need for modernized security and legal strategies to address AI-related threats. Meanwhile, other AI-related developments include Slingshot Insurance emerging from stealth with $27 million in seed funding, and discussions around Anthropic's rapid growth and its implications for the industry.