Sign In

Communications of the ACM

ACM TechNews

White House Worries About Bad AI Coding


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
Binary code.

A report released by the White House this week warns algorithms may have so much power in day-to-day life that ethical frameworks for designing automated computer systems may be needed.

Credit: iStockphoto

The White House released a report this week examining the problems associated with poorly designed systems that increasingly are being used in automated decision-making. The report warns algorithms may have so much power in day-to-day life that it may be important to develop ethical frameworks for designing automated computer systems.

In addition, the report says automated computer systems may need to be transparent for testing and auditing.

Meanwhile, a second effort has been studying the future of algorithms through a series of four workshops held across the U.S. to examine artificial intelligence's (AI) impact on society. "We're increasingly relying on AI to advise decisions and operate physical and virtual machinery--adding to the challenge of predicting and controlling how complex technologies will behave," says the U.S. Federal Trade Commission's Ed Felten.

The federal government will produce an AI report following workshops in Seattle, to be followed by meetings in Washington, D.C., Pittsburgh, and New York City in July.

The most pressing concern is algorithmic systems designed to inadvertently discriminate because of bad design. The report notes such a system also could use a poorly designed matching system or could inadvertently restrict the flow of information.

From Computerworld
View Full Article

 

Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA


 

No entries found