UN launches forum to curb military use of artificial intelligence
The UN Institute for Disarmament Research will convene the first global forum on Thursday to discuss how artificial intelligence is being integrated into weapons systems.
By Felo News Desk · Published
The United Nations announced on Tuesday that the UN Institute for Disarmament Research (UNIDIR) will launch an international forum on Thursday to address the growing military uses of artificial intelligence, according to reports from News18 and The Independent.
What happened
UNIDIR plans to hold the inaugural forum on Thursday, bringing together key state actors and experts to discuss AI‑driven technologies that are increasingly embedded in early‑warning systems, information‑gathering, reconnaissance and nuclear command structures. The initiative is presented as the first dedicated international platform for tackling these issues.
What the reports add
- Both outlets note that the UN press conference took place on Tuesday, October 21, 2026.
- News18 highlighted that UNIDIR’s new Centre of Excellence on AI, Peace and Security – launched this summer – received funding from all five nuclear‑weapon states: the United States, Russia, China, Britain and France.
- The Independent added that the forum will run alongside another UNIDIR effort this week aimed at drafting a voluntary framework for responsible industry behaviour in military AI.
- Both sources mention a recent proposal, made last month, for the United States and China to create an AI‑safety alert mechanism to flag security‑threatening developments.
What was said
“The margin for human judgment is increasingly narrow, including in the nuclear domain,”
Robin Geiss, head of UNIDIR, told the press conference, adding that it is hard to gauge how close the world is to machines operating beyond human control. He said the uncertainty surrounding AI’s role in weapons systems “is the issue we’re trying to resolve here.”
“It’s quite difficult to say at which point we’re actually really losing control,”
Geiss warned, noting AI’s creep into early‑warning, information‑gathering and reconnaissance functions.
“Decisions by companies on design and development determine whether AI‑enabled military capabilities are ultimately reliable, predictable and whether they can operate in line with international law,”
He explained that UNIDIR is working on practical, voluntary guidelines for industry to ensure compliance with international law and human‑rights standards.
How it came about
The forum follows a period of heightened concern over AI’s integration into defence. Earlier this month, OpenAI appointed Sarah Yager as its first Human Rights and Responsible Deployment Lead to oversee military AI use, underscoring the broader debate about responsible deployment of the technology (see Felo coverage). UNIDIR’s new initiative reflects a growing international consensus that AI‑enabled weapons require specific governance mechanisms.
Key facts
- UNIDIR will host the first international forum on military AI on Thursday (news18.com)
- Robin Geiss said the margin for human judgment is narrowing, especially in the nuclear domain (independent.co.uk)
- The forum aims to address AI use in early‑warning, reconnaissance and other defence systems (news18.com)
Sources
- [1] news18.com
- [2] independent.co.uk
