The Plan to Build the Jobs of the Future Is Losing Us the AI Race

There is a growing recognition in Washington that the United States is entering a new era of strategic competition, one defined by advances in artificial intelligence (AI), robotics, and autonomy. Other countries, most notably China, are catching up with U.S. innovation in AI and undercutting U.S. leadership in this critical field. If there is one thing the United States might envy about China’s approach to innovation in AI, perhaps it is the fact that China’s one-party system and strong central leader allow for unity of effort and long-term strategic policymaking. With control of the Senate, House, and White House, Republican policymakers are perfectly positioned to develop and implement a national strategy to strengthen the United States’ lead in AI to stimulate growth, create new AI-enabled jobs, and raise standards of living for working Americans. But to do that, we need to recognize the advantages that have always given us an edge in innovation and strengthen them.

The United States has always had three key advantages in technology and innovation that have made our tech industry the greatest in the world: education, immigration, and investment. The U.S. education system is unequivocally the world’s best. Twenty-six of the top 50 universities in the world are in the United States; only 2 are in China. The United States is also the number one immigration destination in the world, particularly for people with advanced education and skills. The United States is home to more than 20 percent of the world’s migrants, but only 5 percent of the world’s population. And when it comes to investment, the United States has one of the most attractive climates for investment and starting businesses. Because of this, 8 of the top 10 technology companies in the world are U.S. companies, and U.S. companies are leaders in machine learning, AI, and autonomy.

Not only are we not leveraging these advantages to enhance our lead in AI, we seem determined to undermine all of our greatest strengths. In education, at the same time that the Trump administration created a $200 million science, technology, engineering, and mathematics (STEM) education fund, it proposed $9 billion in cuts to the Department of Education, including crucial programs to support students who pursue advanced degrees. Meanwhile, House Republicans included measures in their tax reform proposal that would dramatically raise the cost of higher education and undercut universities’ efforts to recruit leading computer scientists to teach the next generation. These measures include excise taxes on university endowments, which are primarily used to fund research programs, endowed chairs for professors, and tuition assistance, and the elimination of tax deductions and credits for students, which could raise tax burdens for graduate students by as much as 300 percent. It also eliminates tax deductions for employers that contribute to tuition costs for their employees, a powerful disincentive for companies to support continuing education for their employees in a field that is evolving as quickly as AI.

While China is working to incentivize Chinese-born and Chinese-American tech entrepreneurs to return to China, the United States is determinedly pushing them away. Increasing barriers to immigration has been a top priority of both the Trump administration and the Republican Congress. The Trump administration has already cracked down on H1-B visas, making it harder for companies to hire foreign workers for skilled jobs, and has vowed to create more opportunities for Americans to get high-paying jobs by limiting access to those positions for foreigners.

Finally, the administration and some in Congress have announced that they are looking at ways to restrict Chinese investment in U.S. AI companies, citing potential national security concerns. In late 2016, the Department of Defense conducted a study that recommended the United States restrict Chinese investment in AI startups in the United States. The report showed how China is using venture capital to facilitate technology transfer to China and as a window into innovation in the U.S. tech industry. China uses venture investments to identify technologies to copy or acquire and to entice leading researchers and developers to join Chinese companies or move to China. In response, the Trump administration began considering measures to restrict China investment in U.S. AI companies, and Senator John Cornyn (R-TX) announced over the summer that he plans to introduce legislation that would strengthen the Committee on Foreign Investment in the United States (CFIUS) oversight of Chinese investments in AI because of its potential military applications.

These approaches ignore our history and the long-term advantages that made the United States the global leader in tech. Cutting back support for our education system, driving away foreign talent, and restricting investment to cut off China’s access to U.S. innovation will not help us maintain our edge over Chinese competitors; rather, it will ensure that we lose out in the global race for AI. Embracing education and immigration and supporting freedom of investment are not just ideological positions, they have material positive impacts on innovation and competitiveness.

Providing tax incentives to businesses so that they will invest and grow may seem like a great idea, but even if we accept that that is what they will do (and this is hotly debated), we should not cut tax incentives for higher education to pay for it. The jobs of the future will be either building or working with computers, but where will business leaders find educated employees with the necessary skills to fill the tech-enabled jobs they create with their tax windfall? The United States already faces a shortage of computer scientists, and we are going to cut incentives for higher education to fund the creation of more computer science (CS) jobs?

Not only are we reducing opportunities for our own citizens to develop the education and skills to do these jobs, we are making it more difficult for these companies to attract qualified foreign workers. The only winner in all this is China, which will get back more of its highly skilled workers educated in the American education system, while continuing to build up increasingly competitive CS programs at Chinese universities.

Ironically, fueling greater demand for CS jobs while cutting the supply of qualified CS workers in the United States is likely to accelerate the trend toward automation of even the high-end digital jobs that conservatives seek to create. Google is investing in a program called AutoML to create machine learning algorithms that write other machine learning algorithms, a solution to the dramatic shortage of qualified AI experts. While these technologies are likely still years or decades away from widespread deployment, exacerbating the shortage of human workers who can build AI systems will, if anything, accelerate their development and further reduce demand for human labor in many parts of the economy.

Furthermore, at the same time that we are talking up the importance of investing in a new generation of innovative American businesses, we are going to increase oversight and scrutiny of investment in innovative industries and restrict foreign investment in U.S. companies developing cutting-edge technologies. While “reciprocity”—the idea that we should impose the same constraints on Chinese businesses in the United States that they impose on U.S. companies in China—sounds like “fairness” on the surface, it ignores the fact that our open and inviting business environment is a crucial reason that the advanced technologies China wants are developed here.

It also ignores the fact that investments in U.S. tech companies are just one small piece of the Chinese tech transfer machine. The Defense Department report identifies 10 major strategies China uses to transfer U.S. technologies, including industrial espionage and cyber theft, forced joint ventures and intellectual property (IP) disclosure for U.S. companies that invest in China, and encouraging Chinese students educated in the United States to go back and build businesses in China. Of all of these, investing in U.S. companies is arguably the leastconcerning, because at least it can be used to fund new innovation in the United States. And if we restrict Chinese investment in U.S. AI companies, it will only further incentivize them to acquire our technologies through techniques like cyber theft of our IP. Instead of restricting Chinese investment in U.S. tech companies, we should ensure that U.S. entrepreneurs whose technology is coopted by Chinese investors are able to take that Chinese capital and plow it back into new, even more advanced technologies.

We have an opportunity right now to define an economic and international policy agenda that tips the AI race back in our favor. It is a chance to take a few lessons from the Chinese AI policy playbook and teach them a few lessons about the great strengths of the U.S. system. But if we remain wedded to the strategy of creating the jobs of the future at the expense of education, immigration, and investment, we are unlikely to succeed.

William A. Carter is a fellow and deputy director of the Technology Policy Program at the Center for Strategic and International Studies in Washington, D.C.

Commentary is produced by the Center for Strategic and International Studies (CSIS), a private, tax-exempt institution focusing on international public policy issues. Its research is nonpartisan and nonproprietary. CSIS does not take specific policy positions. Accordingly, all views, positions, and conclusions expressed in this publication should be understood to be solely those of the author(s).

© 2017 by the Center for Strategic and International Studies. All rights reserved.