Deploying Big Data to Defend the US?

Data from intelligence, surveillance and reconnaissance (ISR) technologies stream in every second of every day from nearly every corner of the globe, fed by a vast and varied network of data-gathering devices and systems controlled by the United States, including a constellation of satellites, squadrons of drones and other surveillance tools. These platforms generate massive amounts of information; the Navy alone creates a Library of Congress' worth of ISR data every day, but the vast majority of that goes unanalyzed. Other experts cite even more modest figures. "We analyze 0.5 percent or less than 0.5 percent of all the data that's available to us," says Michael Moskal, manager of research programs at Modus Operandi, a company that contracts with the Department of Defense on big-data analysis. "What are we going to do with the other 99.5 percent of the data? Now, however, the DOD is trying to narrow the gap between information and analysis by deploying artificial intelligence to enhance that crucial military state -- situational awareness. The program at the forefront of this effort is Project Maven. Launched in 2017, the initiative established an Algorithmic Warfare Cross-Functional Team to "accelerate DOD's integration of big data and machine learning … [turning] the enormous volume of data available … into actionable intelligence and insights at speed," claims an April 2017 Department of Defense memorandum. Project leaders first focused Maven on drone video, which was inundating analysts with daily terabytes of footage. Before the military turned to AI, "it took a team of analysts working 24 hours a day to exploit only a fraction of one drone's sensor data," Gregory Allen, an adjunct fellow at the Center for a New American Security, writes in the Bulletin of the Atomic Scientists. The Maven team initially aimed to create an algorithm by the end of 2017 to help fight ISIS. That ambitious goal bore fruit in December, when the team deployed an algorithm that could identify objects of interest, like cars. "Eventually, we hope that one analyst will be able to do twice as much work, potentially three times as much, as they're doing now.