Security · 16h ago
AI-discovered bugs rarely weaponized, study finds
VulnCheck reports that fewer than 2% of AI-assisted vulnerability discoveries have been exploited in the wild. The finding challenges claims that frontier AI models give attackers a significant edge. Researchers say the hype around AI-powered hacking may be overblown.
Meridian48 take
The low exploitation rate suggests AI's role in offensive security is still limited, but the gap may narrow as tools improve.
Read the full reporting
AI-found bugs aren't proving any easier to exploit despite the hype →
The Register
ai-securityvulnerability-research