USAF denies running simulation where drone ‘attacked’ operator

It didn’t happen and wasn’t even simulated, says the USAF. An ADF drone is shown above. Image: Defence

Multiple news outlets ran with the story that an AI-controlled USAF drone attacked its own operator during a mission simulation where the operator tried to prevent the drone attacking some hostile targets.

Almost immediately, the US Air Force pushed back against the claim and many of those news outlets are reporting that the source of the story, Colonel Tucker Hamilton, the USAF’s Chief of AI Test and Operations ‘mis-spoke’.

The fact-checking web site FullFact.org, reports Colonel Hamilton’s clarification: “Colonel Hamilton further clarified the mistake, saying: “We’ve never run that experiment, nor would we need to in order to realise that this is a plausible outcome”. He added that the US Air Force has not tested any weaponised AI in the manner described, either in real life or through a simulation.”

The so-called rogue AI drone simulation was a hypothetical “thought experiment” from outside the military, FullFact.org reported.