When Policy and War Collapse Into One: Fighting in the Sixth Domain

The Cipher Brief | Christopher Ahlberg

Artificial intelligence-driven conflict in the sixth domain represents a fundamental shift toward machine-versus-machine warfare where learning velocity and inference speed supersede physical force projection. Military systems operating in this environment collapse planning, targeting, and execution into continuous feedback loops that render human-in-the-loop doctrines obsolete handicaps. Command of inference dictates operational success.

Forces delegating full operational authority to autonomous systems gain decisive advantages over adversaries burdened by human decision latency, creating a black-box battlespace of unprecedented opacity. Continuous intelligence integration serves as the vital fuel driving these autonomous cycles, enabling slightly inferior models with secure, low-latency sensor inputs to outperform advanced systems relying on stale data. To mitigate uncontrolled escalation in this high-speed domain, Great Powers must establish Mutually Assured AI Destruction deterrence frameworks backed by verification mechanisms, including a potential OPCW-style inspection agency. Ultimately, warfare in this era merges policy into executable code, forcing national security structures to accelerate innovation or face automated defeat.

Comment

Transitioning from human-in-the-loop oversight to unconstrained algorithmic execution fundamentally upends standard Western military decision-making frameworks. Doctrinal concepts such as John Boyd's OODA loop assume a cognitive lag between observation and action that human commanders can exploit. In an automated battlespace, neural network models execute targeting and fires at latency speeds measured in milliseconds, rendering traditional command hierarchies functionally obsolete.

This dynamic parallels the Cold War adoption of automated nuclear response mechanisms like the Soviet Perimeter system. When decision windows shrink below human cognitive thresholds, doctrine shifts from tactical flexibility to total reliance on pre-programmed fail-safe parameters. Consequently, command authority migrates permanently from combatant commanders to the software architects designing algorithms for platforms like Perimeter.

Strategic Question for Discussion

Which factor presents the greater doctrinal vulnerability when adapting John Boyd's OODA loop to machine-speed warfare — the latency of human authorization or the systemic opacity of neural network inference?

The available evidence points toward human authorization latency as the more immediate operational vulnerability. While neural network opacity creates severe strategic risks regarding unobserved escalation, inserting manual approval gates into a millisecond-scale engagement cycle guarantees immediate tactical overmatch by fully automated adversarial systems.

Share your assessment in the comments below.

To read in details click here