Yuri Kobzar22:44, 26.09.26

Russians and Americans are trying to prevent an international ban on fully autonomous weapons powered by artificial intelligence.

The United States and Russia are seeking to remove several provisions from an international treaty on the use of artificial intelligence . Specifically, Washington and Moscow oppose the idea of ​​strict human control over the use of autonomous weapons powered by artificial intelligence. This was reported by The Washington Post , citing three people familiar with the closed-door negotiations, as well as documents obtained by the publication.

Negotiations on a corresponding agreement at the UN level took place in Geneva in early September. Hundreds of diplomats worked on a document that could potentially become the first international agreement on the use of lethal autonomous weapons. According to The Washington Post, the negotiations became particularly tense in their final stages.

Over the course of approximately 15 hours, American and Russian representatives made numerous changes to the text. The document removed the requirement that autonomous systems operate “predictably” and “reliably,” as well as the provision requiring ethical considerations when using AI in weapons. Furthermore, the parties eliminated a provision that required human verification of military targets identified by artificial intelligence before launching a strike.

According to the publication’s sources, the closed session was held without cameras or civil society representatives. The American and Russian delegations each brought about ten lawyers—almost twice as many as the other negotiating participants. Due to the rapid pace of changes, smaller delegations, according to sources, were unable to follow the proceedings.

“It was death by a thousand paper cuts,” one source told The Washington Post.

As the publication notes, all of this is happening against the backdrop of the increasingly active use of artificial intelligence in the military. According to The Washington Post, the US military already used the chatbot Anthropic to identify nearly 1,000 targets during the first 24 hours of the war with Iran. Israel also made extensive use of AI during military operations in Gaza.

Human rights activists are warning of the potential consequences of inadequate regulation. “This could mean machines can make life-and-death decisions without human oversight,” noted Verity Coyle of Human Rights Watch.

Meanwhile, the United States continues to review its own regulations regarding autonomous weapons. A document adopted by the Pentagon in 2023 under the Joe Biden administration called for the use of human judgment in the verification of autonomous and semi-autonomous systems and required compliance with the laws of war and relevant international treaties. In June 2026, Donald Trump ordered an update to these regulations within 90 days, but as of late September, the new version had not been released.

At the same time, the UN is attempting to determine the future of international negotiations. In November, countries will meet again in Geneva to discuss the possibility of moving toward a formal mandate or a legally binding treaty.

(C)UNIAN 2026