Who Controls AI on Battlefields - the Military or the Model?

Agentic AI , Artificial Intelligence & Machine Learning , Next-Generation Technologies & Secure Development Who Controls AI on Battlefields - the Military or the Model? Former DoD CIO Beavers on Ethics, Reliability and AI as a National Security Tool Jennifer Lawinski • April 9, 2026     Credit Eligible Get Permission Video Player 00:00 00:00 Leslie Beavers, former CIO, U.S. Department of Defense As artificial intelligence moves into mission-critical defense operations, a new tension is emerging between operational reliability and ethical control. The debate unfolding with AI vendors turns into a deeper issue of who will govern AI behavior in high-stakes battlefield situations, said Leslie Beavers, retired brigadier general and former acting CIO and principal deputy CIO of the U.S. Department of Defense. See Also: Defending Identity in the Age of AI Attacks At the center of the issue is trust. Some AI systems embed value-based guardrails directly into their algorithms, potentially overriding human operators if certain conditions are met, Beavers said. While those safeguards are designed to enforce ethical use, they introduce uncertainty in environments where reliability is non-negotiable. "If I'm running an operation, I need tools that work when I need them to work," Beavers said. This dynamic could reshape how the Department of Defense evaluates vendor risk. Beyond technical capability, alignment on control, predictability and mission reliability is becoming a deciding factor in whether AI systems can be deployed in operational settings. Adopting multiple large language models can also be problematic. While a multi-LLM strategy can improve resilience and reduce blind spots, it introduces additional security, governance and architectural challenges that must be actively managed, she said. In this video interview with ISMG, Beavers also discussed: Challenges related to AI ethics, autonomy and operational reliability; How vendor risk is being redefined in the era of frontier AI; Why multi-model AI strategies both strengthen and complicate defense systems. At the Department of Defense, Beavers led enterprise IT and cybersecurity initiatives across defense environments and now advises on the intersection of AI, national security and digital resilience.