Today’s blog post seeks to focus on, and initiate a discussion about, the current administration’s positions on AI R&D support and public policies. We would like to know SIGAI members’ views on the important areas of concern for AI-related policies.
In December 2016, the Obama administration released a report on Artificial Intelligence, Automation, and the Economy. This report followed the Administration’s previous report, Preparing for the Future of Artificial Intelligence, which recommended that the White House publish a report on the economic impacts of artificial intelligence by the end of 2016. The reports addressed readiness of the United States for a future in which artificial intelligence plays a growing role. The Obama Administration’s views are described in the Roadmap for AI Policy by Ajay Agrawal, Joshua Gans, and Avi Goldfarb in the December 21, 2016, Harvard Business Review. Some reference points from outside the US are Artificial intelligence: an overview for policy-makers from the U.K. and China’s planning for AI.
Miles Brundage and Joanna Bryson argued in August 2016 (see Smart Policies for Artificial Intelligence) that a de facto artificial intelligence policy already exists: “a patchwork of policies impacting the field of AI’s development in myriad ways. The key question related to AI policy, then, is not whether AI should be governed at all, but how it is currently being governed, and how that governance might become more informed, integrated, effective, and anticipatory.”
Some potential implications of AI for society involve the speed of change due to advances in AI technology; loss of individual control and privacy; job destruction due to automation; and the need for laws and public policy on AI technology’s role in the transformation of society. An important point is that, compared to the industrial revolution, AI’s impact is happening much faster and at a much larger scale of use than past technological advances. Organizations need to recognize the likelihood of disruption of operations that will happen whether or not change is intentional and planned.
In our current environment, we need to examine the extent of the new administration’s understanding of AI technology and the need for policies, laws, and planning. So far, not much information is available — from specifics about who will be the head of the National Highway Traffic Safety Agency (NHTSA), the main federal agency that regulates car safety, to the administration’s view of time scales. For example, the administration may take the position that AI will not cause job losses for many decades, which view could distort assumptions about labor market trends and lead to policy mistakes. These views on the future of AI could impact policies that promote programs to promote entrepreneurship and job creation. A few days ago an executive order established the American Technology Council with an initial focus on information technology. The status of the White House Office of Science and Technology Policy is not available on the OSTP Website. AI technology and applications will continue to grow rapidly, but whether or not public policy will keep pace is in doubt.
Please share your ideas via comments to this post and email messages to aimatters@sigai.acm.org.
I had the privilege of attending SAF ART INT June 28, 2016 White House’s Workshop on Safety and Control for Artificial Intelligence co sponsored by the President Obama’s Office of Science and Technology Policy. The workshop was hosted at Carnegie Mellon. Among some of the more lively slightly heated side discussions on AI Regulation were a few vigorous discussions about the part that State’s rights might play in AI regulation. In other words how much AI regulation will the Federal government defer to the States, Governors, Attorney General’s ,etc. In the current attack on science climate The Trump Administration may be much more inclined to off load AI regulation to the States. Do we really want totally decentralized government regulation with respect to AI? Will we have to worry about Red state AI vs Blue state AI ?