WASHINGTON — The U.S. military needs to rapidly adopt artificial intelligence technology to aid in tracking targets, data analysis and managing complex missions, Air Force Secretary Frank Kendall said Dec. 2 at the Reagan National Defense Forum.

Speaking on a panel at the annual forum in Simi Valley, California, Kendall touted AI as an essential technology needed to strengthen U.S. national defense capabilities. 

Although AI has become an integral part of our today’s business across many industries, its use in military systems has raised concerns about unintended outcomes and escalation of an AI arms race.

Kendall argued that when associated with applications such as autonomous weapon systems, AI tends to garner negative connotations. But like other businesses that are leveraging AI to mine data, DoD wants to use AI to enhance human intelligence, to help understand the battlefield and empower human judgment through automating tasks like analyzing sensor data.

“These technologies cannot be stopped,” he said. “It’s being embedded into all of our products, it is being used where it provides a competitive advantage, and the government needs to not get in the way of that.”

The Air Force and the Space Force, Kendall added, need AI tools for data analysis, target recognition, sensor management and a whole host of tasks that currently overwhelm human operators.

Need to ‘gain confidence’

AI should not be viewed as a monolithic technology, he said. “Quite frankly, anybody who’s doing software for anything right now is calling it AI because that’s supposed to be the cool thing,” Kendall added. 

As new applications emerge, he said, “we do need to find ways to evaluate this technology, become confident in it, have the ability to trust it, and to get it into fielded capabilities as quickly as we can.”

There are “enormous possibilities,” Kendall warned, “but it is not anywhere near general human Intelligence equivalents. We’re not talking about that. We’re talking about pattern recognition. We’re talking about enormous efficiency improvements.”

“Our job on the government side more than anything else is to thoroughly understand this technology, have the expertise we need to really get into the details of it,” said Kendall, “and get the confidence we’re going to need to ensure this can be used ethically and reliably when it’s in the hands of our warfighters.”

When pressed about fears regarding the weaponization of AI given its dual-use nature, Kendall affirmed any military applications would comply with international law. These systems will be governed by the laws of armed conflict, just like any other weapon system, and there will be a human in the loop for lethal decisions, he noted.  

‘Battle management’ for space

Kendall insisted that AI technologies will be critical to help military planners understand the battlefield, for example, using Earth observation data from satellites. 

The U.S. and other countries can “watch what the other side is doing in peacetime and have a very good idea of what the threat looks like,” he said. “But what we need is a battle management system for space which keeps track of all those objects, keeps track what the other side is doing, that helps our operators decide with a huge degree of automation how to respond when an attack occurs.”

“If we’re gonna take advantage of AI and the automation that we can get out of AI technologies, we’ve got to put the resources against it, invest in real products,” Kendall said. “That’s the path we’re on.”

Sandra Erwin writes about military space programs, policy, technology and the industry that supports this sector. She has covered the military, the Pentagon, Congress and the defense industry for nearly two decades as editor of NDIA’s National Defense...