AI-Controlled Fighter Jets Defeat Human Pilots in Pentagon Simulations

Thumbnail Image

The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.

In 2020, the US Department of Defense conducted simulated air combat where AI-controlled drones, notably from Heron, defeated top human fighter pilots in all matches. The Pentagon plans to expand such AI-versus-human training exercises in 2024, raising concerns about future military risks from autonomous AI systems.[AI generated]

Why's our monitor labelling this an incident or hazard?

The article explicitly describes AI systems integrated into autonomous combat drones capable of lethal engagement, which have already demonstrated superiority over human pilots in simulated combat. The AI's role in decision-making for targeting and attacking enemy aircraft directly relates to potential injury or death (harm to persons) and disruption of military operations (critical infrastructure). The development and planned deployment of such AI systems constitute a direct AI Incident due to the realized harm in simulation and the imminent real-world application with potential lethal consequences. The article does not merely speculate but reports on concrete tests and military plans, meeting the criteria for an AI Incident rather than a hazard or complementary information.[AI generated]
AI principles
SafetyAccountabilityDemocracy & human autonomyRespect of human rightsTransparency & explainability

Industries
Government, security, and defenceMobility and autonomous vehiclesRobots, sensors, and IT hardware

Affected stakeholders
General public

Harm types
Public interest

Business function:
Research and development

AI system task:
Goal-driven organisationReasoning with knowledge structures/planning


Articles about this incident or hazard