On February 24, 2020, the US Department of Defense formally adopted five ethical principles for the use of artificial intelligence. The principles followed a 15-month process led by the Defense Innovation Board, which consulted AI and technical experts, current and former defense leaders, and the public. The Joint Artificial Intelligence Center, then directed by Air Force Lt. Gen. Jack Shanahan, was given responsibility for putting the principles into practice.
The five principles cover the department’s use of AI in both combat and non-combat settings. They are: Responsible, meaning “DOD personnel will exercise appropriate levels of judgment and care while remaining responsible for the development, deployment and use of AI capabilities”; Equitable, taking “deliberate steps to minimize unintended bias in AI capabilities”; Traceable, with “transparent and auditable methodologies, data sources and design procedures and documentation”; Reliable, with “explicit, well-defined uses” subject to testing across their life cycles; and Governable, including the ability to detect unintended behavior and to “disengage or deactivate deployed systems that demonstrate unintended behavior.”
The principles drew on existing US law, the laws of war, and longstanding defense policy. They did not ban any category of weapon, but they set out a framework the department said its AI work should follow, including the requirement that humans remain responsible and that systems be capable of being switched off.
The 2020 adoption made the United States one of the first militaries to publish a formal set of AI ethics principles. They later became a reference point for the department’s broader AI governance, including the work of the office that succeeded the Joint AI Center.