According to Andrew Hallman, the Deputy Director for Digital Innovation at the CIA, machine learning has tremendously evolved. Hallman said that the Digital Innovation wing at the CIA has significantly improved “anticipatory intelligence.” Anticipatory intelligence, he described, can be used to predict various activities across the globe. It can also be used to predict future social unrest up to five days in advance.
“We have, in some instances, been able to improve our forecast to the point of being able to anticipate the development of social unrest and societal instability some I think as near as three to five days out,” said Hallman
Anticipatory intelligence uses a mesh of algorithms and analytics to predict future events. Deep learning and other forms of machine learning play into this improved anticipatory intelligence. Analysts are able to make connections between entities or events that may not have been apparent without deep learning. Not only do CIA analysts have access to CIA data sets; as of this summer, the digital wing is utilizing open data sets.
In reference to the open data sets, CIA Director John Brennan said:
It’s been a tremendous, I think, advantage as far as trying to fulfill our various missions. We’re trying to deploy more and more of these open source specialists into our mission centers. Having the ability to leverage those – the open source environment and the open source tools — and bring it together with your clandestinely acquired information is just so enriching in terms of how we’re able to understand and create new knowledge.
Several months ago, in July, the CIA announced their intention to improve event prediction. The announcement was made in reference to improving national security. Terrorist attacks were the specific focus of the meeting. The terms “advanced warning” and “early indicators” were used, both in reference to attempted coups in Turkey.
At The Next Tech, Hallman explained that until the recent CIA modernization, policy-makers have had complete confidence in outdated intelligence. He said that it was “much harder to convey confidence for the policymaker who may make an important decision from advanced analytics with deep learning algorithms.”
Hallman explained that the intelligence scene has completely changed. Analysts are “becoming more proficient in articulating” observations to policymakers. Under the new digital directorate, the CIA is fully embracing the digital era.
What this anticipatory intelligence adds up to, Hallman explained, is a clear picture of events that will be unfolding in the future. Machine learning, anticipatory intelligence, and open source data sets have opened a window to the times ahead.