For the Defense Department, last month’s executive order on artificial intelligence was the starting gun, and the department doesn’t mean to lose the race it’s been preparing for for some time. Air Force Lt. Gen. Jack Shanahan, director of the DoD’s new Joint Artificial Intelligence Center, told lawmakers that he’s already trying to stand up a small office around robotic process automation, a specific type of AI aimed towards improving business practices.
“When you talk about smart automation, or in the vernacular of the industry, robotic process automation, it’s not headline grabbing in terms of big AI projects, but it may be where the most efficiencies can be found,” Shanahan said during a March 12 hearing of the Senate Armed Services committee’s subcommittee on Emerging Threats and Capabilities. “That’s the case if you read some of the dailies in industry, whether it’s in medicine or finance, this is where early gains are being realized in AI. Some of the other projects we take on in the department are probably years in the making in return on investment. These other areas I think will be much shorter term in return on investment.”
He said he’s already met with the department’s chief management officer and chief data officer to discuss the matter, which he’s convinced will yield many opportunities to help augment people currently doing those jobs. It’s early in the process; he said the department wasn’t concentrating on RPA a few months ago. But he’s already prioritizing, and he thinks finance will be the first place to apply the new technology.
But RPA is just the tip of the iceberg. Shanahan said that the department has more than 500 AI projects, and JAIC is currently working to evaluate them.
“The number changes depending on who wrote it, but I think in fiscal 2018, the number was 511 projects had AI as their primary focus across the department,” Shanahan said. “The question is: are all those 511 projects toward a common end, in support of the national defense strategy? The question of synchronization is essential to where we’re going in the JAIC.”
He said JAIC wants to ensure all the programs are working toward the same end: furthering the defense of the country. So JAIC is trying to find out what projects are also coming in 2020, and get a handle on their funding. He clarified that he’s not trying to threaten anyone’s budget, just account for the projects and where the money is being spent.
And there’s more projects being started all the time. Peter Highnam, deputy director of the Defense Advanced Research Projects Agency, said DARPA’s new AI Exploration program, which accelerates the rate of awarding contracts for AI, is finding a lot of traction, particularly with schools and small businesses.
“Typically, we put out a call for research proposals, people apply, and six to nine months later, they get selected on contract. We have something called AI Exploration, by which we are driving the research community to explore the space … aggressively,” Highnam said. “We post a topic, and we award a contract within 90 days of posting the topic. We’ve now done this six times. We’ve invested so far on the order of $45 million in this, and there’s tremendous uptake in the research community for these opportunities. All unclassified, all fundamental work.”
And while DARPA is funding AI research and development through this program, the Defense Innovation Unit is trying to cultivate commercial AI projects.
“DIU’s AI portfolio focuses on understanding, tracking and vetting commercial companies’ abilities to solve high-impact problems identified by military leadership,” Michael Brown, DIU director, said. “AI projects today include work with the Air Force, Army, Navy, and components as well as the Joint Chiefs of Staff. As a foundational technology, DIU’s AI portfolio specifically prioritizes projects that address three major impact areas where AI is proven to excel commercially.”
The first of those three areas is computer vision, which means adding automation to object identification and infrastructure assessment. The second is data analytics and predictions, where DIU is currently prototyping predictive maintenance systems for the Air Force and Army that could save billions of dollars. And the last is strategic reasoning, where an AI system could map the probabilities in a chain of events, and formulate alternate strategies.
Building a Culture
Shanahan said another focus of the JAIC is building a culture favorable to AI within the department. Right now, he said DoD is experiencing a combination of top-down pressure to adopt AI, combined with bottom-up innovation. But that innovation needs an outlet, he said. People need to feel free to try and to fail. And the systems need to get into the hands of operators so they can be evaluated, experienced and experimented with.
But the same also goes for the top-down pressure. He said AI capabilities need to be demonstrated to those who are resistant to the concept and think AI is just a matter of science fiction.
But Shanahan is also looking beyond just the department, and beyond even a whole-of-government approach to AI. He said there needs to be a “whole-of-society” perspective to AI, because this will be a multi-generational effort, and there’s currently not enough AI talent to go around.
Brown also touched on this problem.
“How do we build on [the AI executive order] to create common purpose about this being important?” he asked. “My concern would be how many Americans know about the national order on artificial intelligence, and how many young people are we reaching to inspire that this needs to be their mission? Because this technology race, especially on AI, is going to be multi-generational. It’s not going to be lasting through one administration or two. So we’ve got to get the national purpose behind us to support what we can do to leverage that in the Defense Department.”