AI Wins Dogfight: USAF Veteran Loses To Lightweight Software Program [VIDEO]
An advanced, lightweight A.I. known as Alpha, which was developed by Psibernetix, Inc. with assistance from the Air Force Research Laboratory, has successfully beaten all its challengers in simulated dogfights. Alpha's victims included it's A.I. predecessors, and recently, even the most tenured USAF pilots.
In recent simulations, researchers pitted two jets controlled by Alpha against four fighter aircraft running the laboratory's previous A.I. program. Despite being outnumbered by its opponent, Alpha was able to effortlessly shoot down the enemy aircraft. Eventually, Alpha's kill streak became so notable that a more challenging opponent was needed.
Thus, Retired USAF Colonel Gene "Geno" Lee was called. Lee stands as one of the USAF's more tenured pilots and instructor. With thousands of hours of flight experience, the veteran has been involved with the development and training of A.I. programs like Alpha since the 1980s. Just like Alpha, Lee's kill record was also very notable, with the pilot being able to shoot down A.I.-controlled jets with ease.
When faced with Alpha, however, Lee quickly found himself outmatched. During his many matches against the A.I., he never won a single dogfight.
"Until now, an AI opponent simply could not keep up with anything like the real pressure and pace of combat-like. This may be artificial intelligence, but it represents a real challenge," Lee said.
He also remarked that so far, Alpha is "the most aggressive, responsive, dynamic and credible A.I. (he has) seen to date."
Apart from its deadly accuracy and skills, another thing that makes Alpha so fascinating is the fact that the A.I. is so lightweight. During the combat simulations, the developers of Alpha ran the complex program on an entry-level desktop computer worth about $500. Alpha's creators even stated that the A.I. would be able to run on a device that is as lightweight as the Raspberry Pi, one of the world's most basic computers.
A.I. programs have recently been shown to best humans at their programmed fields. Recently, Google's AlphaGo, a Google DeepMind program designed to master the complex game of Go, handily beat the game's human champion. Indeed, it seems like the age of machines is now beginning.