Science

It will soon be too late to control the use of autonomous weapons, warns the Red Cross Weapon technology

The International Committee of the Red Cross and rights groups have warned that it will soon be too late to put the “genie back in the bottle” as the development of autonomous weapons appears to be outstripping self-control efforts.

The stark warning came as more than 120 member states of the UN Convention on Certain Weapons concluded a week-long conference in Geneva.

Despite mounting pressure, an agreement to negotiate an independent arms control instrument over the next two years fell short in a decision rights groups say is needed to deal with the looming threat.

“We are starting to see artificial intelligence … being put on the battlefield and perhaps already in weapons systems. If we leave this unchecked, it will be very difficult to put the genie back in the bottle,” said Dr. Cordula Droege, director of international law, movement and cooperation at the ICRC.

Autonomous weapons are military systems that operate without human supervision, often using artificial intelligence or other forms of robotics.

As artificial intelligence has grown significantly in recent years, militaries and manufacturers have developed highly advanced weapon systems for both offensive and defensive purposes.

These programs include the use of AI in the production of target banks, as was done by the Israeli army in Gaza, and autonomous drones in the Ukraine-Russia war, with “the ability to keep a mile”, where AI takes over the final path of the drone to its target.

The president of the ICRC, Mirjana Spoljaric Egger, and the UN secretary-general, António Guterres, called for a legally binding instrument with clear prohibitions and restrictions on the use of weapons. In particular, the ICRC called for a ban on the use of autonomous weapons against personnel, as well as ensuring that weapons systems behave predictably and placing restrictions on weapons including limiting their use to military purposes.

Experts say that given the rapid development of autonomous weapons and AI, failure to create adequate regulations could have dangerous consequences.

“It’s not about banning all autonomous weapons programs, but putting restrictions on the international agreement of red lines that should not be crossed,” said Droege. “We are very close to crossing those red lines, especially targeting people with machines should be a moral red line.”

Some rights groups have warned that any delay in controlling these weapons would risk allowing private industry and the military to dictate how private weapons are used.

Already, private weapons have led to civilian casualties. Three civilians were killed by a Russian aircraft guided by an AI system in July in Zaporizhzhia in southeastern Ukraine. This unmanned aircraft tried to crash into a gas station, probably intending to shoot it down, but crashed into a wall, killing three people in the rubble.

In Gaza, Israel has used AI-assisted targeting methods to fill its target banks faster than a human since the start of the war. As a result, rights groups have alleged that civilians and people with strong ties to militant groups were killed accidentally.

In Iran, the US strike that attacked a primary school, killing 156 civilians including 120 school children, was reportedly the result of faulty intelligence gathered by AI systems.

Anna Hehir, head of AI military management at the Future of Life Institute, said continued delays in a firm agreement could see “more of the mass casualty cases we’ve seen in the last 18 months to two years”, warning that some of these mass casualty incidents could undercut what has already happened. He said manufacturers and the military have been trying to develop “crowd” capabilities, where hundreds or thousands of autonomous drones operate simultaneously.

“This could be very damaging, just because of the number of programs that can be autonomous. All it takes is for one of them to make an algorithmic error and then the whole crowd is used and it’s down to a method based on that error … especially if it’s programmed to target people.”

There is also the risk of proliferation, as non-state groups increasingly use AI to program their weapons systems, as was the case when the Houthi movement in Yemen used Claude to “vibe code” ballistic missiles.

Armed Islamist groups in Africa’s Sahel region have reportedly used AI tools to improve drone ballistics and operational planning.

“This is the only important thing this week. It’s not really the substance of the laws. What’s on the agenda here, after 13 years, is that the politics will finally be there to start focusing on private arms control,” said a senior official with knowledge of the UN negotiations.


Source link

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button