US Woos Other Nations for Military-AI Ethics Pact

The U.S. will spell out ethics, principles, and practices for the use of artificial intelligence in military contexts in a new declaration Thursday, with the hope of adding cosigners from around the world. The announcement is intended to highlight a "contrast" between the U.S. approach and what one senior defense official called "the more opaque policies of countries like Russia and China." 

U.S. Undersecretary for Arms Control and International Security Bonnie Jenkins will announce the declaration at an AI in warfare conference in the Netherlands. 

“The aim of the political declaration is to promote responsible behavior in the application of AI and autonomy in the military domain, to develop an international consensus around this issue, and put in place measures to increase transparency, communication, and reduce risks of inadvertent conflict and escalation,” Jenkins told Defense One in an email. 

One of the key aspects of the declaration: any state that signs onto it agrees to involve humans in any potential employment of nuclear weapons, a senior State Department official told reporters Wednesday. The declaration will also verbally (but not legally) commit backers to other norms and guidelines on developing and deploying AI in warfare— building off the lengthy ethical guidelines the Defense Department uses. Those principles govern how to build, test, and run AI programs in the military to ensure that the programs work as they are supposed to, and that humans can control them.  

The UN is already discussing the use of lethal autonomy in warfare. But that discussion only touches a very small and specific aspect of how AI will transform militaries around the world. The U.S. government now sees a chance to rally other nations to agree to norms affecting other military uses of AI, including things like data collection, development, test and evaluation, and continuous monitoring.

The State Department and the Pentagon are also hoping to attract backers beyond their usual partners. 

“We would like to expand that to go out to a much broader set of countries and begin getting international buy-in, not just a NATO buy-in, but in Asia, buy-in from countries in Latin America,” the State Department official said. “We're looking for countries around the world to start discussing this…so they understand the implications of the development and military use of AI…Many of them will think ‘Oh this is just a great power competition issue,’ when really there are implications for the entire international community.”

While the declaration doesn’t specifically address how nations that adopt it will operate more effectively with the United States military, it does align with sentiments expressed by the Biden administration, most notably National Security Advisor Jake Sullivan, on the need for countries with shared democratic values to also align around technology policy to build stronger bonds. 

The senior Defense official told reporters: “We think that we have an opportunity to get ahead in a way and establish strong norms of responsible behavior now…which can be helpful down the road for all states and are consistent with the commitments we've made to international humanitarian law and the law of war. Neither China nor Russia have stated publicly what procedures they're implementing to ensure that their military AI systems operate safely responsibly and as intended.”