New International Forum on Military AI
The United Nations is launching a new initiative to address the military risks of artificial intelligence, with a specific focus on the potential for machines to act outside of human control. The effort is led by the U.N. Institute for Disarmament Research (UNIDIR) and aims to bring together key players from across the globe to develop frameworks for responsible behavior and ensure compliance with international law.
On Thursday, UNIDIR will launch a new forum as part of its Center of Excellence on AI, Peace and Security. The initiative is designed to bring together actors from every relevant sector and all over the world to work on AI in military uses.
Robin Geiss, head of UNIDIR, was vague on exactly what the specific actions of the new initiative will entail but stated it will be part of other AI efforts being launched by the center that week. The U.N. agency plans to hold consultations this week with defense technology companies, software developers, subcontractors, investors, governments, civil society, academia, and experts.
Warnings on Human Judgment and Control
During a U.N. press conference on Tuesday, Geiss warned that "the margin for human judgment is increasingly narrow, including in the nuclear domain." He stated it is difficult to assess how close the world is to machines acting outside of human control.
Geiss noted that AI is creeping into early warning systems and reconnaissance, making it "quite difficult to say at which point we're actually really losing control." He pointed to a proposal last month for the U.S. and China to set up an AI safety alert system for issues that could threaten national security as evidence of major powers prompting action.
According to Geiss, there was previously widespread reluctance to deal with AI guardrails, but a growing realization has led to action on military AI.
Industry Frameworks and Armed Group Risks
Another initiative associated with the center will focus on the role of industry. This effort aims to develop a framework for responsible industry behavior with practical, voluntary guidelines that follow international law and ensure human rights.
UNIDIR is also launching an initiative focused on the risks posed by AI misuse by armed groups. Geiss warned that such actors are already using the technology to support operational planning, execution, and logistics.
Funding and Global Context
The Center of Excellence on AI, Peace and Security, which was established this summer, received funding from the five major nuclear powers: the U.S., Russia, China, Britain, and France.
Risks posed by AI were a major topic in speeches by leaders of many of the U.N.'s 193 member nations at their annual high-level meeting last month.