White House Weighs In on Artificial Intelligence

The Internet Law Resource Center™ is the complete information solution for practitioners in cyberlaw. Follow the latest developments on ICANN’s gTLD program, keyword advertising, online privacy,...

By Michaela Ross

Oct. 12 — The White House released two reports Oct. 12 laying out potential policy issues and recommendations surrounding artificial intelligence technologies and outlining federal funding priorities to foster them.

Future regulation of artificial intelligence (AI) should encourage innovation while protecting public safety and economic and job growth, the White House said in its reports. One of the reports described the current state of artificial intelligence and made a series of recommendations to monitor and stimulate development of the field. In the second report, the administration set priorities for federally-funded artificial intelligence research with a focus on building a skilled workforce.

The government, industry and others should “work together to develop the positive aspects of the technology, manage its risks and challenges, and ensure that everyone has the opportunity to help in building an AI-enhanced society,” White House officials wrote in an Oct. 12 blog post.

Artificial intelligence, which includes technologies that exhibit human problem-solving skills or learning capabilities such as those found in Amazon.com Inc.'s personal assistant Echo and Uber Technologies Inc.'s autonomous vehicles, is a booming field. The reports listed cybersecurity, ethics, privacy, economic inclusivity and fairness issues while stopping short of calling for heavy regulation. That approach could increase the cost of compliance or slow the adoption of beneficial innovations, one of the reports said.

The White House's light-handed strategy might mean new regulations won't happen for a long time, if ever, Scott Weingaertner, partner at White & Case LLP, told Bloomberg BNA. The legal framework for artificial intelligence uses may be shaped through litigation instead, he said.

“You can see the possibility of things going the wrong way, people discovering that after the fact and, if there’s not regulation, it will be left to the courts to deal with,” Weingaertner said.

Products and services using artificial intelligence technologies have surged in recent years with improvements in the availability of data, improved capabilities in computing power and advances in AI algorithms combining with investments from highly competitive tech companies, said Eric Horvitz, co-chair of the recently launched non-profit Partnership on Artificial Intelligence (AI) to Benefit People and Society. The industry group includes Google’s DeepMind, Facebook Inc., IBM Corp., Amazon and Microsoft Corp.

Rising Stakes

As AI capabilities improve, applications in high-stakes areas including self-driving cars, healthcare and military systems mean failures may lead to costly and deadly outcomes, Horvitz said. There are also concerns about job losses with the increasing competency of AI systems, he said.

A light touch approach may be best, but regulation could be appropriate in some “safety-critical areas,” Horvitz said. It may be valuable to require certain AI technologies to be tested and rolled out in several phases with tracking, reporting and disclosures of performance, he said.

“No one wants to see premature regulation and premature legal activity get in the way of innovation, especially in the early days of the rolling out of technologies into the world where they promise to be quite beneficial,” Horvitz told Bloomberg BNA. “However, we need to also work to do our best to field these advances in a careful and thoughtful way.”

Next Steps

The administration's reports cover broader social implications related to AI, but are light on concrete next steps and dates that will spur accountability, Weingaertner said.

One of the reports recommends a follow-up study on automation and the economy to be released before the end of the administration in the coming months.

The White House Office of Science and Technology Policy hosted a series of public-outreach activities this year, including workshops and a request for information to inform its recommendations.

The White House issued the reports on the eve of its Frontiers Conference in Pittsburgh Oct. 13. The technology-focused event will feature several panels discussing machine learning, robotics and automation best practices and challenges.

To contact the reporter on this story: Michaela Ross in Washington at mross@bna.com

To contact the editor responsible for this story: Keith Perine at kperine@bna.com

For More Information

The complete text of the “Preparing for the Future of Artificial Intelligence” report is available at http://src.bna.com/jjY.

The complete text of the “National Artificial Intelligence Research and Development Strategic Plan” report is available at http://src.bna.com/jj1.

Copyright © 2016 The Bureau of National Affairs, Inc. All Rights Reserved.