Participants at a summit on the use of artificial intelligence (AI) in the military fields Tuesday called for maintaining human control for reliable military use of AI and preventing its use from spurring the proliferation of weapons of mass destruction.

The calls were made in a joint declaration adopted following the two-day Responsible AI in the Military Domain (REAIM) conference, hosted by South Korea, a multilateral gathering for discussions on building the governance for the use of AI in the military fields.

“Appropriate human involvement needs to be maintained in the development, deployment and use of AI in the military domain, including appropriate measures that relate to human judgment and control over the use of force,” the “Blueprint for Action” declaration, released by the foreign ministry, showed.

“We stress the need to prevent AI technologies from being used to contribute to the proliferation of weapons of mass destruction, and emphasize that AI technologies support and do not hinder disarmam
ent, arms control and nonproliferation efforts,” it said.

This year’s REAIM conference brought together some 2,000 government officials and private experts from 90 countries to explore ways to address the military use of AI in terms of benefits as well as potential risks.

Russia, in its prolonged war with Ukraine, was not invited to this year’s meeting, just as it was not last year.

Of the participating states, 61 countries endorsed the declaration. More countries could be added to the list, including China, which is reportedly considering endorsing this year’s summit outcome, according to a diplomatic source.

The meeting gained attention as risks of the misuse of AI have raised serious concerns, such as the potential for an arms race.

The declaration stressed that humans remain responsible and accountable for the use and effects of AI applications in the military domain.

The importance of human involvement should apply to “all actions critical to informing and executing sovereign decisions concerning n
uclear weapons employment, without prejudice to the ultimate goal of a world free of nuclear weapons,” it said.

The participants in the declaration called for establishing appropriate safeguards to make sure the AI applications are reliable and trustworthy, and reduce the risks of malfunctions or “unintended consequences,” such as those from data, algorithmic and other biases.

It also noted that possible high-impact applications in the military domain deserve policy attention, including AI-enabled weapons, AI-enabled decision-support systems for combat or cyber operations.

In the closing ceremony, Foreign Minister Cho Tae-yul called for efforts to put the principles into practice, saying that it is people’s “will and choices” that will determine the future.

“This is only the beginning. We need to further develop the blueprint and work to implement measures as we keep pace with the speed of technological advancements,” Cho said in the closing remarks.

This year’s REAIM conference was co-hosted by the Neth
erlands, Singapore, Kenya and Britain. The inaugural session took place in The Hague in February last year.

Source: Yonhap News Agency