'Disaster for humanity': Experts to RT on joint AI project by Google & Pentagon

Published time: 7 Apr, 2018 13:06Edited time: 8 Apr, 2018 08:38

Get short URL

© U.S. Air Force

• 
  

Hundreds of Google employees are up in arms over the company's partnership with the Pentagon in AI technology, fearing it may be used for war. Experts told RT the "questionable" alliance could result in "disaster for humanity."

Google employees wrote a letter to the company's CEO, Sundar Pichai, calling on the US tech giant to immediately pull out of a controversial program that many fear could be used for warfare.

Read more

All-Seeing Eye: Google working with Pentagon on using AI for drone improvement

"We believe that Google should not be in the business of war," the letter obtained by The New York Times and published earlier this week stated. 

Gizmodo broke the news about Google's partnership with the US Department of Defense (DoD) last month, adding that Project Maven, whose stated mission is to "accelerate DoD's integration of big data and machine learning," was established in April 2017. The project will see Google developing AI surveillance to help the US military scrutinize video footage captured by US government drones "to detect vehicles and other objects, track their motions, and provide results to the Department of Defense."

Google claims that the technology is human-friendly and is actually designed to "save lives" and "scoped to be for non-offensive purposes." But Noel Sharkey, Emeritus professor of AI at Sheffield University, told RT that the fears of Google employees "are correct."

The Maven program "is all about bringing AI to the immediate conflict zone" he argued, adding that Google may simply be too naïve here about the real use of its technology.

"Once you start working with the military, you have no control over what they use your product for, and that's very worrying," Professor Sharkey said.

He cautioned that while drones now have human operators, which are at least "looking at the target, engaging with the target and trying to calculate its legitimacy," things can take a drastic turn.

"If Google's imagery is very good, they will stop using that operator, allow robots to go out on their own, find their own targets and kill them without human intervention. And this is a disaster for humanity."