The Agentic Olympus: The “Gods” of AI in Irregular Cognitive Warfare

Irregular Warfare Center  |  John A. Pennell, Elise G. Annett, James Giordano

Russian and Iranian cognitive warfare operations below the threshold of kinetic conflict leverage artificial intelligence, deepfakes, and automated narrative distribution to target NATO alliance cohesion, democratic decision-making, and public trust. To counter these high-tempo gray-zone campaigns, future military command structures must transition toward an "Agentic Olympus" framework that synchronizes specialized machine intelligences across command-and-control enterprises.

This distributed decision architecture links real-time intelligence, cyber defense, logistics, and narrative monitoring systems to maintain operational cognitive coherence under continuous adversarial pressure. Adversaries utilize reflexive control, information exhaustion, and narrative saturation to degrade shared situational awareness and manipulate an opponent's decision calculus before formal engagement occurs. Within this evolving battlespace, successful command depends on integrating computational scale, predictive reasoning, and automated information processing while preserving human ethical responsibility, strategic intent, and institutional authority. Operational success in modern irregular warfare ultimately relies on governing distributed cognitive ecosystems to protect decision integrity against synthetic deception and rapid narrative manipulation.

Comment

Integrating multi-agent autonomous decision software into command architectures shifts the vulnerability of command and control from network bandwidth to data integrity. Ukraine's Delta situational awareness system demonstrates how distributed information feeds enable rapid targeting cycles, yet also exposes nodes to sensor spoofing and automated data poisoning. When algorithmic agents filter raw sensor feeds before human evaluation, adversary cognitive manipulation targets the verification algorithms rather than human operators.

During NATO's Exercise Steadfast Defender 2024, multi-domain command nodes highlighted the operational friction caused by asynchronous AI recommendation tempos across coalition networks. Synthetic narrative injection during such exercises accelerates decision paralysis by forcing automated verification sub-agents into recursive validation loops. Consequently, the operational resilience of platforms like the Delta network depends on establishing deterministic fallback protocols when agentic synthesis breaks down under cognitive EW.

Strategic Question for Discussion

Which carries more weight in preserving command integrity during high-tempo cognitive warfare — hardening sensor verification pipelines within platforms like the Delta network, or establishing human-in-the-loop overrides when automated verification loops stall?

The trajectory of AI-enabled command architectures suggests that hardening automated sensor verification offers greater resilience against high-speed data poisoning than manual intervention alone. Relying on human overrides during dense cognitive EW attacks risks creating cognitive bottlenecks that negate the speed advantage of systems like the Delta platform. Consequently, operational survival points toward embedding deterministic cryptographic validation directly within agentic software layers to maintain decision tempo.

Share your assessment in the comments below.

To read in details click here