LET’S NOT TALK: USAF DENIES AI KILLED DRONE OPERATOR IN SIMULATION EXERCISE  An Air Force official has denied staging a simulation exercise where an artificial intelligence drone went rogue and killed its human operator. The US military is attempting to use AI for everything from fighting fires to operating fighter jets, but a deadly glitch would be the stuff of horror movies.

The controversy sprang from comments made by Colonel Tucker ‘Cinco’ Hamilton, the Chief of AI Test and Operations, USAF. He was addressing an audience at a conference in London by the Royal Aeronautical Society on May 23-24. Col.Hamilton said that he was involved in flight tests of autonomous systems, including robot F-16s that can dogfight, and that when a weaponized AI system is given the autonomy to act on its own it could develop “highly unexpected strategies” to achieve its goals.

Col. Hamilton claimed that in a simulated test, an AI-enabled drone was given the task of destroying surface-to-air missile sites and that it realized that the decisions made by its human operator were interfering with its higher mission - to kill the Surface to Air Missiles. He said that the system decided to wipe out the operator because it saw no advantage in returning the decision to a higher authority. But on Friday, the Air Force issued a statement denying that it ever ran such an experiment. It said that Col. Hamilton misspoke’ during his presentation and that the rogue drone story had been taken out of context.

Rick Wiles, Doc Burkhart. Airdate 6/02/23