AI Risks in Federal Cybersecurity Survey Findings

This technical report presents findings from the 2026 Booz Allen Velocity Survey, focusing on AI risks in federal cybersecurity. The survey involved over 100 federal cyber and IT leaders, exploring their concerns regarding AI-powered attacks and the readiness of their agencies to deploy AI technologies securely. A significant majority, 79%, expressed high levels of concern about adversaries weaponizing AI for cyberattacks in the near future. The report details various AI-enabled threats, including adversarial AI and AI-accelerated vulnerability exploitation, and assesses the perceived readiness of agencies to implement AI-powered defenses. It also highlights barriers to adopting these technologies, such as integration challenges and budget constraints. The findings indicate a dual challenge for federal agencies: defending against AI threats while ensuring the secure deployment of their own AI systems. The report concludes with a call for a coordinated approach to address the gaps in governance, technology, and talent necessary for effective AI integration.