Overwhelmed by thousands of hours of Islamic State surveillance tapes, the Pentagon is catching up to the future by looking to machine learning and big data to help its military surveillance analysts.

Codenamed Project Maven, an Algorithmic Warfare Cross Functional Team will sift through hours of video caught by drones flying over Iraq and Syria in the hope that it can discover actionable intelligence. Over the last few years The Pentagon has deployed a large number of drones over areas its thought IS could be hiding, but according to the Pentagon, military and civilian intelligence operatives in the U.S. are no match for the sheer size of the content.

“We have to tackle the problem a different way,” Air Force Lt. Gen. John N.T. Shanahan, director for defense intelligence for warfighter support, told Defense One. “We’re not going to solve it by throwing more people at the problem. That’s the last thing that we actually want to do. We want to be smarter about what we’re doing.”

Automating video analysis might give operatives the chance to do something other than the “mundane” work of recording unusual activity on a spreadsheet, said Shanahan. The algorithmic team will be headed by Shanahan, who told Defense One he has spent a lot of time lately in Silicon Valley looking for the best engineers and technology to assist the U.S. military in surveillance automation.

“There are a thousand things we want to do with artificial intelligence, machine learning, deep learning, computer vision, but everybody has cautioned us, don’t take on too much the first time you do this,” Shanahan said, adding that the tech industry was “light years” ahead of the Department of Defense.

In one example of the mundanity of the work analysts face, around 60 percent of video captured by military drones is basically commute time as the drone heads to an area of interest. Other footage could be hours of bad weather. Nonetheless, analysts are expected to watch all the content. What the military hopes is that by automating the process, software can go through thousands of hours of footage in no time at all and find all the important bits, which will then be further inspected by a human eye.

U.S. military officials reportedly view this kind for automation as indispensable in the near future, although where it will work, the military is still pondering. “You have to go after a manageable problem,” said Shanahan. “Solve it, show early wins and then start to open Pandora’s box and go after all of these other challenges.”

Image: AK Rockefeller via Flickr