
Mozilla reported that an AI model identified 271 zero-day vulnerabilities in a recent Firefox release, demonstrating how rapidly AI can surface many flaws. That capability increases pressure on vendors to remediate issues quickly and on disclosure processes to keep pace. The recent dispute between Microsoft and a researcher who published exploit code adds to patterns of public disclosures and leak-driven risk, highlighting the need for clearer norms and faster coordination between researchers and vendors as AI expands vulnerability discovery.
Click a connection line between nodes to view confidence and evidence.