Did an AI-controlled drone really kill a human? Get the point!

Recently, an official from the United States Air Force (USAF) Tested with a drone controlled by Artificial Intelligence (AI). Until then, very good.

The detail is that during the simulation, could have killed the device operator. The statement was made by Colonel Tucker Hamilton, USAF head of AI operations and testing.

Now, according to the US website Deputy, Hamilton backed out in his claim, saying he had “underexpressed himself” about the situation. and that no one died during the test. Incidentally, such a simulation would never actually happen.

Drone killed operator in experiment

According to an e-mail to the press agent, everything was a “hypothetical experiment” based on real scenarios and probabilities, not a real simulation with real people. So the operator died by “lying”.

“We’ve never done this experiment, and we won’t have to do it to figure out what the result will be,” Hamilton said in an email to the US. DeputyIt was formalized by the Royal Aeronautical Society (RAeS), where the Colonel made his keynote presentation saying that a drone had taken someone’s life. “While this is a hypothetical example, it illustrates the real challenges AI tools face and why the Air Force is committed to the ethical development of AI.”

Before the dispute, Hamilton talked about Artificial Intelligence and highlighted how technology can transform society. “We must face a world where artificial intelligence is already here and is already transforming our society. […] It makes it easy to deceive and manipulate. We need to develop ways to make AI more robust and be more aware of why software makes certain decisions.”

“Artificial intelligence is a tool we must shape to transform our nations… or if misused, we are doomed,” the Colonel concluded.