A now-fixed Zoom screen-sharing flaw let someone on a call take over another participant’s device, according to researchers cited by WIRED. The striking detail is how quickly they say the bug was found: fewer than 20 prompts to a public AI tool.
The case shows how AI can change vulnerability research. A model that helps reason through code paths, interface behavior, or edge cases can make skilled security work faster. That can help defenders find bugs sooner, but it can also lower the effort required for attackers to discover exploitable mistakes.
Zoom has fixed the issue, so the immediate risk described in the report is addressed. The wider lesson is not that AI discovered the bug alone, but that human researchers can use general AI tools as accelerators. Security teams may need to assume that flaws in widely used collaboration software will be found more quickly, and that patch speed, responsible disclosure, and defensive testing will matter even more as AI-assisted research becomes routine.