Retired United States Air Force Colonel Gene Lee was consistently defeated by an artificial intelligence fighter pilot program in simulated air combat.
Get ready for a new wave of Skynet fears, this time prompted by an artificial intelligence (AI) fighter pilot program that has overwhelmingly dominated other AI air combat systems and a series of human experts in various hyper-realistic simulations.
News of the AI’s achievement was published in the Journal of Defense Management in March and recently announced by the University of Cincinnati’s UC Magazine. The school is the alma mater of Nick Ernest, a 2015 UC College of Engineering and Applied Science doctoral graduate and the current president and CEO of the AI development company, Psibernetix, Inc.
As a contractor to the United States Air Force Research Laboratory the firm created ALPHA, an AI air combat system that not only consistently surpassed previous AI programs, but has also repeatedly defeated a string of expert human fighter pilots including retired United States Air Force Colonel, Gene Lee.
An instructor and expert in aerial combat with extensive fighter aircraft experience, Lee has been dogfighting AI opponents in simulations for more than three decades and has advanced training in air combat strategy and tactics, but neither his experience or knowledge proved good enough to beat the AI.
“Not only was Lee not able to score a kill against ALPHA after repeated attempts, he was shot out of the air every time during protracted engagements in the simulator,” UC Magazine reports.
“I was surprised at how aware and reactive it was,” Lee told the publication. “It seemed to be aware of my intentions and reacting instantly to my changes in flight and my missile deployment. It knew how to defeat the shot I was taking. It moved instantly between defensive and offensive actions as needed.”
Air combat is extremely technical. Flying at speeds up to 1,500 miles per hour or more, fighter pilots must make split-second decisions that can mean the difference between life or death.
In this regard, the speed at which ALPHA can make precise decisions is stunning.
According to UC Magazine, “ALPHA can take in the entirety of sensor data, organize it, create a complete mapping of a combat scenario and make or change combat decisions for a flight of four fighter aircraft in less than a millisecond. Basically, the AI is so fast that it could consider and coordinate the best tactical plan and precise responses, within a dynamic environment, over 250 times faster than ALPHA’s human opponents could blink.”
DON'T MISS: Watch Out World: Robots Can Evolve
Perhaps the most impressive of ALPHA’s achievements is the minimal computing power that is required to run the advanced AI.
According to UC Magazine, in ALPHA’s air battles with Lee the program relied on the same amount of computing power found in $35 Raspberry Pi computers. The latest version of the AI runs on the computing power of a $500 PC.
From left, retired United States Air Force Colonel Gene Lee, Psibernetix President and CEO Nick Ernest and Programming Lead David Carroll. Credit: Lisa Ventre, UC Creative Services
Beyond the AI’s current use in training and research, the system could be deployed in the future as a virtual or unmanned wingman for human fighter pilots, according to Kelly Cohen, a UC aerospace professor who helped develop ALPHA.
“ALPHA would be an extremely easy AI to cooperate with and have as a teammate,” Cohen told UC Magazine. “ALPHA could continuously determine the optimal ways to perform tasks commanded by its manned wingman, as well as provide tactical and situational advice to the rest of its flight.”
With unmanned aerial vehicles like drones rapidly taking over the skies, and Psibernetix’s plans to continue to develop and improve on ALPHA, it doesn't seem too far-fetched to envision the future of increasingly sophisticated air combat as a human-less theater of war.
“This may be artificial intelligence, but it represents a real challenge,” said Lee.
Read this next: Computer Beats Human at the World’s Most Complex Board Game