AI Models Are Great at Finding Security Bugs. Can They Tell When They're Fixed?

AI models have shown effectiveness in identifying security bugs, but their ability to determine when these bugs have been fixed is still uncertain. The capability of AI models to detect security issues is well-established, however, verifying the resolution of these issues is a different matter. This raises questions about the limitations and potential applications of AI in cybersecurity. The role of AI in bug detection versus bug fixing is a topic of interest. AI models are proficient at finding security bugs, but their capability to confirm fixes is less clear.