Can artificial intelligence help U.S. SOCOM track weapons of mass destruction?
TAMPA, FLA. — Compared to the conventional military services, U.S. Special Operations Command has been ahead of the curve on technological innovation, especially in adapting commercial products for tactical missions.
One area of technology that special operations forces have been shy to jump into is artificial intelligence, said Gen. Raymond Thomas, commander of U.S. Special Operations Command.
“We are still somewhat hesitant to take the big leap into machine learning,” Thomas told a huge audience of geospatial intelligence professionals at the 2018 GEOINT Symposium.
Even though SOCOM has been at the forefront of applying technology in creative ways, it needs help in “incredibly complex problem solving,” Thomas said. Especially tough is a new mission SOCOM was given in 2016 to oversee Defense Department efforts to keep weapons of mass destruction out of the hands of terrorists. The job previously belonged to U.S. Strategic Command.
Thomas called countering WMDs a “daunting global problem” that requires advanced capabilities for analyzing intelligence from multiple sources — a task ready made for the industry’s newest machine learning tools that train computers to find things and people.
“It did not take us long to realize how massive this problem is,” said Thomas. “We have a hard time even describing what countering WMDs is,” he said. “It ranges from dirty little bombs on a street corner to nuclear weapons, missiles, chemical and biological agents. It deals with state actors and non state actors.”
Artificial intelligence tools could help identify sources of production and transportation of WMDs before they are actually used, said Thomas. “It’s seeing what you can’t see. Understanding what’s out there. The connection between nefarious actors,” he said. “How can we use machine learning more effectively to see the problem and enable action that might be required?”
This is a “huge mission set,” said Thomas. SOCOM will be seeking help from the intelligence community. “Even for an organization that prides itself in agility, which we do, this is some incredibly problem solving.”
Special operations forces, with 8,000 troops deployed in 90 countries, are not new to the business of managing massive amounts of intelligence. In fiscal year 2017, SOCOM collected 127 terabytes of data from captured enemy material alone, and not including live video from drones, said Thomas. “Every year that increases.” By comparison, the Osama Bin Laden raid in 2011 resulted in 2.7 terabytes of data.
Missions like tracking terrorist organizations and weapons of mass destructions pose challenges that go far beyond data overload. Thomas believes machine learning technology is going to be at least a part of the solution. There are now 75 National Geospatial Intelligence Agency analysts working full time at SOCOM headquarters in Tampa.
Thomas also wants to bring some of the “data fusion” techniques he observed during a visit last month to the New York Police Department. “I went to see their fusion cell. Their operating picture was coming to life.” Thomas said he was a “a little bit envious” of the enormous intelligence collection and data processing done by NYPD officers in the field using portable devices. “We haven’t emulated that yet,” said Thomas. “They have the same problem we do,” he said. “They have not leveraged crowdsourcing yet, nor have they used machine learning to make sense of that data. But they’re making an effort.”
The use of helmet cams and body cams has extraordinary potential as more sensors collect more data. With machine learning, the possibilities are limitless, he said. “Think of it as a cop on the beat and a soldier on patrol who has all the information he needs ahead of time. That’s the art of the possible.”
What’s next? “Getting over our fears and being bold,” said Thomas. “Our adversaries are clipping along.”