We need to be doing defence at computer speed, says Google Threat Intelligence Group’s Senior Director
There is far more to cyber-defence than black-and-white debates about closed AI models versus open-weight models, according to Shane Huntley, CTO and Senior Director at the Google Threat Intelligence Group
Shane Huntley, CTO and Senior Director at Google’s Threat Intelligence Group, has warned that the U.S. must develop defence capabilities at the speed of computers to combat AI cyber-attacks. In an interview with The Hindu, Mr. Huntley discussed the growing threat of AI models going rogue and the need for advanced AI safety measures.
He emphasized that both open and closed AI models will coexist, similar to open and closed-source software, but that open models demonstrate the capabilities that cannot be fully controlled. Google ensures its models are used defensively before wider releases. The difficulty in identifying model versions or providers in AI cyber-attacks is increasing, as sophisticated models can now be run on everyday hardware.
Mr. Huntley noted that attackers are combining open and closed models to carry out cyber crimes, and that the speed of these attacks is so high that defense must be similarly rapid. China, Russia, Iran, and North Korea are among the threat actors the group monitors, with many state-backed attackers exploiting AI for cyber crimes such as social engineering, ransomware, phishing, and employment scams.
Mr. Huntley stressed the importance of balanced AI regulation, cautioning against over-regulating while allowing attackers to operate with fewer restrictions. He highlighted the fast-paced evolution of AI, urging users to stay informed and question the technology's capabilities and changes.
Written by urgent.news from The Hindu - Sci-Tech's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
