DeepMind To Do Deep Thinking On A.I. Ethics

DeepMind launches Ethics and Society program

Seeks to delve into ethical issues related to A.I. and robots

DeepMind is a London based artificial intelligence research lab and a world leader in A.I. It was acquired by Google in 2014 and is a part of the Googe Alphabet group and has been at the cutting edge of developing algorithms that can teach themselves to play very sophisticated games.

DeepMind developed the algorithm that was responsible for crushing the world's best player in the game of GO. It is continuing to work on many of the most difficult technical problems in developing computer algorithms, as it plays a key role in making A.I. smarter, more capable, and more autonomous.

Now, it is diving into the softer sciences by launching a brand new research unit, DeepMind's Ethics & Society. It is designed to "explore the real world impacts of A.I." And it has a dual aim "of helping technologists put ethics into practice and to help society anticipate and direct the impact of A.I. so that it works for the benefit of all."

It plans on being a fellowship by inviting leaders in Humanities and Social Sciences to contribute to the discussion. One hot-button issue is the potential bias in computer algorithms risk assessing criminals. A study conducted a year ago believes it found racial bias: in risk assessment of the potential for criminal activity, the study showed higher potential skewed more toward blacks. Source

But it seems there are more pressing problems when available historical data already indicate that blacks are more likely to commit crimes compared to other racial groups - those are just facts. Now, if A.I. can determine which individuals are likely to commit a crime like in Minority Report, no matter the race of the person, which it most likely will, then the ethical issue becomes one of privacy.

The one ethical issue that seems more prevalent, affecting much more people, is the economic impact A.I. and robots will have on society through job loss. Prioritizing ethical issues based on overall impact is an ethical consideration in itself. Maybe the algorithms will point this out.