The United States is urging China and Russia to follow the example of the U.S. and other nations and declare that only humans, and not artificial intelligence, would make decisions regarding the use of nuclear weapons.
Paul Dean, principal deputy assistant secretary in the Bureau of Arms Control, Deterrence, and Stability at the State Department said in an online briefing that the U.S. had made a "clear and strong commitment" that only humans would have control over nuclear weapons. He added that France and Great Britain had made similar commitments.
"We would welcome a similar statement by China and the Russian Federation," said Dean. "We think it is an extremely important norm of responsible behaviour and we think it is something that would be very welcome in a P5 context." The "P5" are the five permanent members of the UN Security Council.
The State Department has said that U.S. Secretary of State Antony Blinken and Chinese Foreign Minister Wang Yi spoke briefly about artificial intelligence during their meeting last week.
"I do think that there is a real opportunity right now as countries increasingly turn to artificial intelligence to establish what the rules of responsible and stabilizing behavior will look like. And I think it’s – this is a conversation... that I think all major militaries and major economies – like the United States and China – have to deal with," Dean said Thursday.
He mentioned that the U.S. and 54 partners – not including China and Russia – have endorsed a political declaration on responsible uses of military AI, which will "ensure there is no accountability gap in the use of artificial intelligence in the military, and ensure that the applications are designed and used according to rigorous technical specifications, with some designs built in to ensure that there can be safeguards and that the technology can be used in a responsible way."
"This technology will really revolutionize militaries across a range of applications," Dean also said.
"And I would emphasize here that the issue is not limited to battlefield use but these technologies will be used by militaries across the entire range of their operations on efficiencies and logistics, decision making. And I think this presents great promise and I think there’s significant upside in this, but of course, as with all new technologies, there are risks if the technology is not used in a responsible way."
I doubt very much that any such agreement will stop any military from using AI any way they want if it means the difference between victory and defeat.
The issue of AI not "controlling" the use of nuclear weapons is curious considering that computers already remove almost the entire human element from the decision to use nukes. While humans are still giving "go," or "no go" orders, all the information that leads to that decision is sifted through several scenarios, possibilities, and outcomes after being run through some of the most sophisticated software ever created. The difference between AI-controlled systems and what we have today is really a question of speed.
First, the AI security dilemma features the possibility of extraordinarily fast technological breakthroughs, incentivizing states in competition with peers in AI technology to move first rather than risk being second. For example, the U.S. National Security Commission on AI found that “defending against AI-capable adversaries [notably China] operating at machine speeds without employing AI is an invitation to disaster.”
Second, the AI security dilemma risks placing latent offensive capabilities in civilian hands, such as the massive data facilitated by communication and navigation satellites. Whereas the traditional security dilemma is driven primarily by the misinterpretation of defensive military capabilities, the AI security dilemma also can be driven by the misinterpretation of ostensibly peaceful commercial capabilities
The AI race is already underway. Finishing second is not an option.
