"AI vs AI: New algorithm automatically bypasses your best cybersecurity defenses"
Researchers at Endgame security firm have demonstrated the use of artificial intelligence to modify malware code in order to circumvent anti-malware machine learning within antivirus software. The experiment performed by researchers brings attention to the issue of all AI having blind spots that could be exploited by other AI to bypass security. Though researchers were only able to bypass security measures of antivirus software 16 percent of the time in the experiment, this circumvention still poses a significant threat as a system can be infected by malware that has been modified by AI to become undetectable. This article further discusses this demonstration by researchers and what it means for machine learning in security.
Submitted by Anonymous
on