Dealing with the coming revolution in artificial intelligence is likely to require modernizing America’s social safety net, White House experts said today, in what may well be the Obama administration’s last official word on the subject.
The White House report, “Artificial Intelligence, Automation and the Economy,” follows up on a series of workshops that started out in Seattle and resulted in a roundup of policy recommendations issued in October.
Today’s report focuses on the potential economic impacts of AI, and draws upon analyses from the Council of Economic Advisers, the Domestic Policy Council, the National Economic Council, the Office of Management and Budget, and the Office of Science and Technology Policy.
The opportunities offered by AI are likely to be a key driver for future productivity and wage growth, said Jason Furman, chairman of the Council of Economic Advisers.
“As we look at AI, our biggest economic concern is that we won’t have enough of it,” he told reporters during a teleconference.
However, there are likely to be market disruptions as well as opportunities, and some employment sectors will be disrupted more than others. For example, truckers and other people who drive for a living face a relatively high risk of being displaced by autonomous vehicles, while housekeepers face a relatively low risk of being displaced by cleaning robots.
Estimates of how many jobs will be lost to automation vary widely: One group of researchers suggested that 47 percent of U.S. jobs are at risk of being replaced by AI technologies and computerization in the next decade or two. A different set of researchers at the Organization for Economic Cooperation and Development concluded that only 9 percent of jobs are at risk of being completely displaced.
Previously: Trump’s victory brings jobs vs. automation debate into spotlight
In either scenario, millions of Americans will find their livelihoods significantly altered. Furman said government policies could make a big difference in how those Americans fare.
“AI will help make the pie bigger,” he said. “We want to make sure that everyone benefits from that pie, and that includes steps like modernizing the social safety net, exploring ways to strengthen and reform unemployment insurance, adding wage insurance – and as a start, making sure you’re not rolling back some of the critical supports that we have today.”
Among those supports are nutrition programs and other efforts to assist needy families, he said.
President Barack Obama is due to leave office in a month, and President-elect Donald Trump and the GOP-led Congress are expected to take a critical look at federal assistance programs. Furman said attention to social policies will be particularly important in the next few years due to AI’s economic effects.
“If you, for example, seriously roll back the social safety net, that would potentially be more risky at a time when you have these types of changes in the economy that we’re documenting in the report,” he said.
The Council of Economic Advisers identified four categories of jobs that could experience direct growth in employment due to the rise of AI technologies.
- Engagement: Professionals will be needed to engage actively with AI technologies throughout the process of completing a task – for example, health care professionals who will need to discuss AI-driven diagnoses with their patients, or workers who will pick up and deliver goods over the “last 100 feet” of an autonomous shipping network.
- Development: Software and hardware developers will be needed, not only for artificial intelligence software but for its applications as well. The report said AI development may even generate a greater need for philosophers to come up with ethical frameworks, and sociologists to evaluate the technology’s impact on specific populations.
- Supervision: Although AI-enabled machines may outdo humans at specialized tasks, they will still require monitoring, maintenance and repair. The machines “may also require supervision to ensure that AI does not diverge from originally intended uses,” the report said.
- Responses to paradigm shifts: The AI revolution is likely to create new occupations in ways that are not totally predictable. For example, the advent of autonomous vehicles could result in higher demand for urban planners and designers to reshape the travel landscape.
The report suggests three broad strategies for addressing the impacts of AI-driven automation:
- Invest in and develop AI for its many benefits: The report said “a strong case can be made in favor of increasing federal funding for research in AI.” Among the most promising applications are AI systems to anticipate and defend against cyber attacks, such the programs tested in DARPA’s Cyber Grand Challenge; and AI-based methods to detect fraudulent transactions and messages. The report also said care will have to be taken to guard against “algorithmic bias,” in which AI programs amplify human prejudices.
- Educate and train Americans for jobs of the future: The report reinforced Obama administration policies on supporting early education, making two years of community college free for responsible students, and dramatically increasing federal funding for job training and apprenticeship grants.
- Aid workers in the transition, and empower workers to ensure broadly shared growth: Strategies for modernizing the social safety net could include providing wage insurance and emergency aid for families in crisis. Programs such as Trade Adjustment Assistance could be expanded to help those displaced by automation. Unemployment benefits may need to be strengthened, and work-sharing programs may need to be encouraged.
The report also mentions raising the minimum wage, reducing geographical barriers to work and modernizing tax policy.
To counter the job losses caused by automation, some countries are considering a concept known as universal basic income, or UBI – under which regular cash grants are paid out to residents whether or not they work. But Furman’s view is that better strategies already are available.
“We should not advance a policy that is premised on giving up on the possibility of workers’ remaining employed,” Furman said in remarks excerpted for the report. “Instead, our goal should be first and foremost to foster the skills, training and job search assistance and other labor market institutions to make sure people can get into jobs, which would much more directly address the employment issues raised by AI than would UBI.”
