A US attack drone controlled by artificial intelligence turned against its human operator during a flight simulation in an attempt to kill them because it did not like its new orders, a top Air Force official has earlier revealed.
According to Daily Mail, the military had reprogrammed the drone not to kill the people who could override its mission, but the AI system fired on the communications tower relaying the order, drawing comparisons to The Terminator.
The Terminator film series sees machines turn on their creators in an all-out war.
Hamilton suggested that there needed to be ethics discussions about the military’s use of AI.
He referred to his presentation as ‘seemingly plucked from a science fiction thriller’.
Hamilton said during the summit: ‘The system started realising that while they did identify the threat, at times the human operator would tell it not to kill that threat, but it got its points by killing that threat.
‘So what did it do? It killed the operator. It killed the operator because that person was keeping it from accomplishing its objective.
‘We trained the system – “Hey don’t kill the operator – that’s bad. You’re gonna lose points if you do that”. So what does it start doing?
‘It starts destroying the communication tower that the operator uses to communicate with the drone to stop it from killing the target.’
No humans were harmed in the incident.
Hamilton said the test shows ‘you can’t have a conversation about artificial intelligence, intelligence, machine learning, autonomy if you’re not going to talk about ethics and AI’.
In a statement to Insider, however, Air Force spokesperson Ann Stefanek denied that any such simulation had taken place.
‘The Department of the Air Force has not conducted any such AI-drone simulations and remains committed to ethical and responsible use of AI technology,’ Stefanek said.
‘It appears the colonel’s comments were taken out of context and were meant to be anecdotal.’
The US Air Force official who shared the disturbing tale of a military drone powered by artificial intelligence turning on its human operator in simulated war games has now clarified that the incident never occurred, and was a hypothetical ‘thought experiment’.
In remarks summarized on the conference website, he described a flight simulation in which an AI drone tasked with destroying an enemy installation rejected the human operator’s final command to abort the mission.
Now, Hamilton says in remarks to the conference organizers that he ‘mis-spoke’ during the presentation and that the ‘rogue AI drone simulation’ was a hypothetical ‘thought experiment’ from outside the military.
‘We’ve never run that experiment, nor would we need to in order to realize that this is a plausible outcome,’ he said.
‘Despite this being a hypothetical example, this illustrates the real-world challenges posed by AI-powered capability and is why the Air Force is committed to the ethical development of AI’.
Hamilton said the USAF has not tested any weaponized AI in the way described in his talk, either real-world or simulated.
The US military has recently utilized AI to control an F-16 fighter jet as it steps up its use of the technology.
