Google Under Fire

PUBLISHED: 12:24 AM 8 Mar 2018

Google Under Fire For Assisting Defense Department With Artificial Intelligence For Drones

The AI will help the military analyze drone footage.

Apparently, the partnered project – which is titled Project Maven- has the goal to create advanced technology that can pinpoint and even determine what particular objects are in the drone footage.

Yesterday, tech outlet Gizmodo revealed that Google is working with the U.S. Department of Defense in order to enhance the military drones’ capabilities by making them artificially intelligent.

Apparently, the partnered project – which is titled Project Maven– has the goal to create advanced technology that can pinpoint and even determine what particular objects are in the drone footage. Gizmodo also noted that a Google spokeswoman said that the company is working “to develop policies and safeguards” around the use of using machine learning for military purposes.

Naturally, this private-public initiative was supposed to be secret and even classified for the most part. However, citing anonymous sources, Gizmodo pointed out that pertinent information was widely circulated last week through an “internal mailing list.”

What seems quite interesting is the fact that the sharing among some Google staffers was of indignation instead of appreciation or even positive intrigue.

Of course, this doesn’t come as a big surprise, not only because of the political differences but also because of the fact that military drones are a military tool most often used for ultimately violent operations.

As everyone knows, these violent endeavors became extremely famous during the Obama administration, and eventually represented some of the few elements that a significant part of the left always criticized about this period.

According to tech experts, this is actually pretty good news considering that this partnered project could better help identify what a drone is viewing or targeting.

This way, it could conceivably make the warfare weapon and its application in the battlefield way more meticulous than ever and prevent major accidents or unnecessary casualties.

Despite these interesting possibilities, staffers at Google are reportedly still quite worried about the powers and implications of machine learning, as reported by Gizmodo.

One of the most significant moments to explain this move was when Eric Schmidt was appointed the chairman of a larger DOD program called the Defense Innovation Board (DIB) two years ago. As everyone knows, this man is the former CEO of Alphabet, which is the parent company of Google.

According to the Daily Caller, there were plenty of reasons why Schmidt decided to remain on at the helm, despite the rise of President Donald Trump.

Naturally, almost everyone at Silicon Valley thought he was going to resign, not only because of the fact that he has always been a ferocious critic of the Republican, but also a staunch supporter of the Democratic Party.

While he since stepped down as the chief executive of the tech conglomerate, he remains involved in the DIB and as one of the most important board members of Alphabet.

According to the Google spokeswoman, the tech company provides its TensorFlow application programming interfaces (APIs) to the Project Maven.

APIs are software-based rules that let computer programs communicate in the most efficient way. On its part, TensorFlow is a very popular set of APIs and many other tools for Artificial Intelligence (AI) capabilities such as computer vision and machine learning.

Apparently, the feature is part of a recent Pentagon contract involving Google’s cloud unit. This one is trying to wrest more government spending from cloud-computing leaders Microsoft Corp and Amazon.com Inc.

Back in August, U.S. Defense Secretary James Mattis decided to visit Google headquarters in Mountain View, California, and met with some company executives.

Basically, the reason for this meeting was to discuss the best ways to use cloud computing, AI, and cybersecurity for the Pentagon.

At a meeting in July, Google’s board recommended that the Defense Department look at ways to take the vast data that exists in the enterprise and turn it into something that could be “actionable.”

According to minutes of this meeting, Schmidt said that every single piece of data should be stored somewhere no matter what the structure is. That’s because “we” could always go back and then discover the structure and eventually use it in the most favorable way.

Regarding this issue, Google executive Milo Medin warned of a tremendous lost opportunity when the Pentagon data isn’t collected.

Tech experts believe that’s especially true in AI, which requires huge amounts of information to train software algorithms that improve automatically.

Moreover, Medin said that every single fighter plan or destroyer that returns from a certain mission or operation and doesn’t provide the data it collected, represents a serious loss of capability in machine learning and training that is forever lost.

So far, it hasn’t been reported any resign from a Google staffer.