(TMU) — According to two U.S. military researchers, the United States needs a “Dead Hand” that would ensure an automated, rapid response to any nuclear threat.
Their solution? To fit the U.S. nuclear arsenal with artificial intelligence (AI) controls.
In a column for military blog War on the Rocks, U.S. nuclear warfare experts Adam Lowther and Curtis McGiffin suggest that AI is precisely the solution for what they see as a lagging U.S. response system to threats.
The two men make the claim that the current NC3 system is dependent on outdated Cold War technology, posing the danger that it “may be too slow for the president to make a considered decision and transmit orders” in case of a nuclear attack on the United States.
The authors are well aware that such a proposed system invokes the horrifying apocalyptic imagery of U.S. science fiction films, noting:
“Admittedly, such a suggestion will generate comparisons to Dr. Strangelove’s doomsday machine, War Games’ War Operation Plan Response, and the Terminator’s Skynet, but the prophetic imagery of these science fiction films is quickly becoming reality.“
And while the Cold War may long be a thing of the past, experts claim that such a system “that might seem unfathomable” remains necessary “to reinforce the desired deterrent effect” of the U.S. war machine in the face of rivals’ nuclear modernization programs while giving a boost to second-strike capabilities.
The potential application of artificial intelligence tech to nuclear warfare is hardly a fresh concept. The United Nations University explains that during the height of late 20th century nuclear rivalry between the U.S. and the former USSR, both countries’ respective military leaderships explored the option of bolstering their nuclear capabilities with varying degrees of AI.
However, the Soviet Union was the only country to develop a semi-automated nuclear launch mechanism to ensure an all-out nuclear response in case their leadership was decapitated in the form of the Dead Hand system.
However, the authors of the article suggest that their hypothetical AI nuclear system would be superior to Dead Hand because “the system itself would determine the response based on its own assessment of the inbound threat,“ meaning that a nuclear explosion wouldn’t be necessary before the system launched an attack.
Experts have warned, however, that AI could increase the risk of grave miscalculations and the destabilization of international relations.
In a 2018 RAND Corporation report, the think-tank argues:
“In view of the capabilities that AI may be expected to enable and how adversaries may perceive them, AI has the potential to exacerbate emerging challenges to nuclear strategic stability by the year 2040 even with only modest rates of technical progress.”
Indeed, a “live hand” has in the past been the only thing preventing tensions from cascading into an all-out nuclear conflict.
In 1983, Soviet Lt. Col. Stanislav Petrov was monitoring radar data when he saw what appeared to be a U.S. missile on a trajectory toward the USSR.
A 1999 report from the Washington Post detailed the event:
“The alarms went off. On the panel in front (of) him was a red pulsating button. One word flashed: ‘Start’.”
Luckily for all of us, Petrov had a hunch that the apparent “missile” was a mere technical error, which proved to be the case.
“I had a funny feeling in my gut.
I didn’t want to make a mistake. I made a decision, and that was it.”