AI & Zero Trust: New Cybersecurity Challenges

Artificial intelligence is challenging traditional cybersecurity approaches, particularly the concept of zero trust. Organizations need to identify how AI is used, where it accesses data, and how systems interact to build strong security. Extending identity-based access controls to AI systems is key for consistent security. Monitoring AI activity and keeping audit records provides visibility and accountability for data handling. Applying uniform policies across all environments, whether cloud, on-premises, or edge, helps reduce complexity. A major concern is that 44% of U.S. workers use unauthorized AI tools, and 46% have uploaded sensitive company information to public AI platforms. Experts emphasize that securing the data powering AI is fundamental. Protecting the AI models alone is not enough; the underlying data must be safeguarded. This shift means zero trust must now protect not just users and devices, but also AI models, prompts, retrieval systems, and AI agents. This is important because unsecured AI can increase risks to security, compliance, and data exposure.