White House: Benefits, inevitability of AI may outweigh potential risks

Artificial intelligence and automation technologies have the potential to alter millions of jobs, yet the positive benefits associated with increased productivity are worth pursuing, according to a December 2016 White House report. The authors – which include staff from the Council of Economic Advisers, Domestic Policy Council, National Economic Council, Office of Management and Budget, and Office of Science and Technology Policy – suggest that responding to the economic effects of this technology will be an important consideration for the next administration, and as a result, policymakers should be prepared for a wide variety of outcomes. Artificial Intelligence, Automation, and the Economy is a follow up to the White House’s October 2016 report Preparing for the Future of Artificial Intelligence.

With regard to AI impacts on the economy, the authors state that it is difficult to provide precise predictions based on the current evidence available. While the strength and timing of these potential impacts are unknown, in general the authors anticipate five economic effects: increased aggregate productivity growth; higher demand for high-level technical skills, and overall changes in the types of skills needed by employers; an uneven distribution of impacts (across sectors, wage levels, education levels, job types, and geographies); significant job market churn, with jobs disappearing and created; and, short- and potentially long-term loss of jobs for some workers. While the White House is careful to not make estimates or projections on jobs lost or created due to automation, a July 2016 analysis by McKinsey & Co. finds that currently demonstrated technologies could automate 45 percent of the activities people are paid to perform, and that about 60 percent of all occupations could see 30 percent or more of their activities automated.

To respond to these challenges, the report concludes with three broad policy strategies. First, the authors state because of the many benefits to AI, policymakers should continue to invest and develop the technologies. Public outreach conducted by the OSTP found that many business leaders, technologists, and economists were interested in more government investment in AI technology, and that a strong case can be made for increased funding for AI research. In October 2016, the White House also released a National Artificial Intelligence Research and Development Strategic Plan.

Second, as the nature of work and skills changes as a result of artificial intelligence and automation, there is a need to prepare American workers accordingly. Impactful investments in high-quality investment at all levels of education, with an emphasis on STEM and college-/career-readiness, present one opportunity to develop the future workforce. The report also identifies the need for a larger, more diverse technical workforce, with the authors stating that “the importance of including individuals from diverse backgrounds, experiences, and identities, especially women and members of racial and ethnic groups traditionally underrepresented in STEM, is one of the most critical and high-priority challenges for computer science and AI.”

Finally, the authors note the need to strengthen the government safety net as a way to assist workers that have, are, or will be transitioning out of employment. Government programs such as unemployment insurance in particular, but also Medicaid, Supplemental Nutrition Assistance Program (SNAP), and Temporary Assistance for Needy Families (TANF), are potential targets for modernization, according to the report. As a way to empower workers to share in the effects of the productivity growth tied to automation, the authors recommend improved employment services that provide guidance to transitioning workers. This may reduce the time these workers spend unemployed, as well as offer meaningful information suited to their particular skills and circumstance.

The relationship between advancements in automation and the ensuing job churn will remain an important part of the political discourse in coming years. While United Technologies ultimately decided to keep some Carrier jobs in Indiana, the CEO also stated they would be investing in automation at the plants to keep them cost competitive – which would lead to fewer jobs. Because private sector research is more likely to be spent on profit maximization, not job retention, continued investments in automation are likely as companies like United Technologies and Ford invest in domestic manufacturing.

Just as the outcomes of artificial intelligence remain shrouded in speculation, so do the potential policy responses from the Trump administration – though expanded social assistance programs for transitioning workers are unlikely, according to an analysis by The Washington Post.