AWS says its AI security agent fixes 89% of bugs — but there’s a catch
AWS made Continuum for code vulnerabilities generally available on Thursday. The service finds and validates bugs in your repositories, checks The post AWS says its AI security agent fixes 89% of bugs — but there’s a catch appeared first on The New Stack .
AWS announced its AI-powered security tool, Continuum, is generally available. The service automatically finds and validates code vulnerabilities, then proposes fixes through pull requests. Continuum demonstrated an impressive 89.0% end-to-end success rate in testing, outperforming previous AI agents. AWS CEO Matt Garman praised AI as a transformative force for security defenders.
However, there may be a catch: Continuum's patch fixed only 37.8% of historical vulnerabilities, despite finding and stopping the crash 89.6% of the time. The benchmark used to evaluate Continuum only tests C and C++ memory-safety bugs, not covering the full spectrum of vulnerability types. Continuum's performance may improve with more time, as AWS reported a 93.7% success rate when given an extended 90-minute window.
The benchmark and testing tools are publicly available, but AWS is the only one showcasing the tool's capabilities so far.
Written by urgent.news from The New Stack's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
