JUST IN: Mastering Human-Machine Learning Crucial for Air Force, Official Says

ntegrating human-machine learning and artificial intelligence into Air Force operations is crucial for maintaining a strategic advantage on the modern battlefield, the service's chief said.

While warfare and national defense are “a human endeavor,” AI and machine learning can and should be used to enhance and improve the capabilities and operation of both human warfighters and machines, Gen. David Allvin, Air Force chief of staff, said at a talk hosted by the Brookings Institution on Feb. 28.

“AI-enabled technology and autonomy certainly aren’t going to replace [humans], but the [service] that augments them and the one that masters the human-machine teaming, I think, is going to have a critical advantage going forward in warfare,” he said.

Leveraging the capabilities of machine learning in order to enable operation to the maximum extent also means allowing the technology to express what can be done on the human side, and that concept is not untapped per se, but “lightly tapped,” Allvin said.

“It has so much potential in the future, from the business systems that help us to make everyday decisions better, to those AI-enabled algorithms that are developed and enable us to develop situational awareness and decision speed,” he added.

AI and machine learning have an important role to play in enhancing situational awareness and decision-making in the Air Force, Allvin said.

“Essentially, if you sort of fast forward into the future, and you start thinking about the algorithms that help situational awareness, that [help] one side make a decision faster than the other, it’s a fascinating thought experiment,” he said. “It’s an interesting thing to think about [the role] algorithms will play in [the decision-making] of senior leaders because that's really where it’s going to happen.”

As for how the Air Force is currently utilizing human-machine learning and AI, Allvin used the example of the collaborative combat aircraft, uncrewed platforms designed to fly and operate alongside crewed aircraft. He said the service is not pursuing the program in a “20th century way.”

With the collaborative combat aircraft, the Air Force is not following a linear development pattern. Instead of developing the aircraft step by step, the service is “doing it all at once,” with the help of both autonomous integration and human involvement, he said.

“At the same time, we’re looking at the platforms. We’re also developing, looking at [and] experimenting with what autonomy can and can't do to help shape how we would use that,” Allvin said. “And at the same time, we’re having an experimental operations unit understand how one might leverage that and integrate it into our Air Force.”

Leveraging and mastering human-machine integration and AI will play a large role in strategic deterrence and maintaining a strategic advantage, he said.

“I believe that the manner in which we develop our force and get the skill sets that we need to be able to leverage advanced technology, human-machine teaming, all those things that I believe are going to be [beneficial] to a future Air Force, getting the talent in place [and] the incentives to retain that talent is going to be key to actually be able to leverage all the technologies and stay ahead of the competitors,” Allvin said.