AI safety warnings mount as frontier models test new limits of cybersecurity - Baltimore Sun

Industry leaders and researchers have warned improving AI models could pose growing cybersecurity risks that require stronger defenses amid ...