Analysis finds AI safety guardrails create cyber defense gaps

Frontier AI lab policies restricting cyber-offensive capabilities limit the tools available to security defenders. This creates an asymmetry where defensive systems lack the parity needed to counter sophisticated AI-driven threats.