The CIA is Probing Artificial Intelligence for A Super Spy Job

The good religious book claims that we will reach a time where, all citizens of the earth will be traceable by a number, which will encompass a universal code “666.” And those who’ll not want to register, a system, (maybe manned by the CIA) will be able to instantly track them…and the prophecy goes on.  Well, could artificial intelligence be paving the way for that now?

Because in the near future, the US’s CIA (Central Intelligence Agency) will be able to know everything that’s happening in a target place, country or region in real time without contact on the ground. What does that suggest to be precise? Simple;

CIA’s Next Super Spy Could Be None Other Than AI

Source: verge

That may sound technical, but you know what, already technologies that point to that possibility are making waves. Google is now using satellite imagery to help the aviation, and defense industries, in decision-making. Facebook is also on record for successfully sharing its satellite data with machine learning.

In a different approach, Alibaba is using a face and sound recognition technology to monitor over 10 million pigs with utter accuracy.  Besides that, China’s police have also deployed a system working with ET Brain that can autonomously understand suspicious activities in CCTV clips.

Now, taking a rather special approach, the US’s CIA is currently pursuing nearly 140 projects all aimed at increasing the job of artificial intelligence in surveillance. Its role will be scanning or perusing through videotapes collected from spy and closed-circuit cameras.

Although in a different niche, something close to this is happening. Where scientists were able to devise an algorithm that turns poachers to pray, by reporting their near real-time actions, that is the AI analyzes live videos taken by hovering drones, to alert the game rangers. Connecting this to human security issues, it’s evident that machine intelligence is extending a helping hand.

Digital Surveillance

Source: cdn

Working as the Deputy Director of CIA’s research team (Science and Technology Division,) Dawn Meyerriecks, is mentioned by CNN, explaining how digital surveillance has been successful that it’s now replacing physical tracking. “Led by Singapore, both wireless-based and closed-circuit television digital surveillance has proved very effective in over 30 countries now, the US needs to advance its monitoring tech from such a foundation.”

Meyerriecks also told how a team of experts used unclassified aerial footages of a street – and paired that with machine learning to generate algorithms that can map out strategic virtual placement of cameras in remote capitals, that are not safe for physical access. What this means in simple language, is that we can have surveillance software linked to a satellite and posing as a map of cameras over a target region without a ground contact.

Besides being super effective and highly manageable, because this can be monitored from a single room, the method can also be used to counter-surveillance – which ideally means updating field agents when, and in case they are being surveyed. This way the US will as well be able to guarantee the security of its diplomats in rival countries like Russia or South Korea, as well as reduce the risk of exposing American spies who end up being kidnaped or killed.

Boosting Surveillance Efficiency

The traditional method has been we only revisit the CCTV footages after a crime has already occurred or if lucky enough, pay attention to what is being recorded when something fishy has been noticed. Well, The National Geospatial-Intelligence manages to move a bit beyond that, Robert Cardillo the agency’s head says he sends part of the security team to spend time analyzing the TV monitors, as part of ensuring security. But, it’s obvious, that is not as reliable or efficient as it ought to be still –the team might sleep on the job.

Mr. Cardillo supports the idea of using AI to analyze what’s going on in a scene because the machine is destined to remain active, compared to humans. Top of that, the algorithms have proved that they can analyze, to detect patterns of events in videos, and alert the security team about suspicious images or moves automatically, for further action.