AI Security Startup Bay Launches with $50M to Vet Agent Actions

Israeli startup Bay AI Inc. has officially launched with $50 million in funding to address the security risks posed by autonomous AI agents. Founded by veterans of the IDF’s Unit 8200, the company aims to move beyond simple prompt filtering by vetting the actual actions an agent attempts to execute. The platform tracks agents across various environments—including laptops, cloud workloads, and SaaS applications—and uses a centralized policy engine called 'Lighthouse' to evaluate risks in milliseconds. If an action is deemed too dangerous, such as unauthorized database deletion, the system blocks it before execution. The company emphasizes that its security checks occur on the machine itself, ensuring low latency and high context awareness. By analyzing the previous 30 days of agent activity, Bay creates tailored security policies for different teams, reducing the burden on security staff. With engineering and product operations based in Tel Aviv, the company is positioning itself to secure the growing enterprise adoption of autonomous AI agents, ensuring that these systems operate within safe, predefined boundaries.