AI‑discovered bugs still hard to exploit, study finds

A new study examines the practical impact of bugs identified by artificial intelligence tools. Researchers compared AI‑generated vulnerability reports with traditional exploit development. The findings indicate that most AI‑found bugs remain difficult to weaponize. Exploit complexity did not decrease in line with the hype surrounding AI security tools. Analysts note that human expertise is still required to turn findings into functional exploits. The report highlights gaps between automated detection and real‑world attack readiness. Industry observers caution against over‑reliance on AI for rapid exploit creation. Future work may focus on bridging detection and exploitation phases to meet expectations.