Among the cutting-edge technologies being employed by public health experts to map various aspects of COVID-19 both at home and abroad, artificial intelligence (AI) faces a test under life-and-death circumstances. The ability of AI systems to undertake pattern detection and predict the spread of the pandemic and its treatments is promising. The benefit of machine learning includes its powerful ability to analyze historic data to find key variables. This task is dependent upon humans, however, specifically in the ability of data scientists who can work on creating data sets that supercomputers then can model. On a global basis, this will require pooling both technical and human resources.
Given the unprecedented nature of COVID-19, historic data inputted for AI analysis may be of limited value. Real-time data comparing growth curves in countries around the world, along with population and demographic information by neighborhood, may prove to be a better vein for producing actionable data anywhere and everywhere. Automated machine learning also may improve the efficiency of data scientists, enabling them to focus on new data generation while relying on computer-to-computer analysis of massive-scale number crunching.
According to the Center for Data Innovation, as of last summer, the United States and China are emerging as the two country competitors for global AI leadership. China’s aspirations are explicit: It seeks to achieve AI dominance both to expand its economic footprint and to fortify its military power.
The Center’s analysis indicates that so far, the U.S. is leading the race, in areas including talent, research and development (R&D), and hardware. One critical aspect for the U.S. is innovation, an area that benefits from robust private equity and venture funding capability. Another is America’s significant strength both in traditional semiconductors and AI-specific computer chips. China, on the other hand, is leading in AI system adoption and in AI data production, which also are important assets.
COVID-19 presents an opportunity to demonstrate in practice the U.S. AI development strengths, highlighted above. It also will reveal longer-term issues regarding how the U.S. may fall behind in AI development if it does not bulk up in its stronger advantages and shore up its weaker competitive areas during this crisis. Federal stimulus funding to help address pandemic concerns could increase incentives for AI R&D, for example. This could take the form of direct subsidies for focused R&D, tax credits, or some combination of these.
Money alone will not be sufficient. We also must be mindful that better domestic AI talent must be developed at America’s great research universities, with more academic research funding necessary to enhance an enlarged homegrown AI talent pipeline.
The Center’s report is correct in noting that “the race to develop or adopt AI is not a zero sum game,” and that “many AI advancements, particularly those focused on health … can benefit all countries. For example, the development of AI systems that can identify diseases faster and more accurately than clinicians, or produce new medical treatments, offers potentially global benefits.”
Competition aside, the larger common task the U.S. shares with China may be more important now than our ongoing trade disputes, which have resulted in the imposition of tariffs on select goods imported by each country. U.S. and China AI players have an opportunity to pool technical resources and know-how, particularly since the strengths of both countries complement one another.
Collectively, the U.S. and China also are in a good position to learn from advancements in the EU, including the work of data scientists in Italy, who have access to unique data based on the rapid spread of COVID-19 there. That’s because much AI research is shared openly already. And the possibility of limited, royalty-free cross-licensing of AI patents held by the U.S. and China may be especially beneficial to bring in much-needed technical capability on a short-term basis.
The ability to compete and cooperate simultaneously can achieve win-win outcomes both sooner and later. The United States should send a clear signal to China now – that it is open to explore AI working arrangements to not only benefit the economies and strategic interests of both countries, but also help achieve the global goal of successfully confronting a worldwide pandemic with the best AI tools that can be brought to bear.
Stuart Brotman is a Fellow at the Woodrow Wilson International Center for Scholars in Washington, based in its Science and Technology Innovation Program. He is also a member of The Media Institute’s First Amendment Advisory Council. This article appeared in Columbia University’s Journal of International Affairs on April 10, 2020.