Source: Yamashita et al., “Measuring the AI content of government-funded R&D projects,” OECD Science, Technology and Industry Working Papers, June 2021.
Commentary: R&D investments from federal agencies such as the Defense Advanced Research Projects Agency (DARPA), the National Institutes of Health (NIH), and the National Science Foundation (NSF) have played a critical role in spurring essential AI innovations that have also driven private AI-related R&D investment. Tracking these public investments, OECD economists produce new findings on the AI-related content of R&D investments made by 13 agencies across 8 OECD countries. Their research finds that total AI-related government R&D funding between all 13 agencies grew from $207 million in 2001 to nearly $3.6 billion in 2019, a 17-fold increase.
In the United States alone, a similar trend holds. The OECD analyzes NIH and NSF’s contact-investment data and finds that, between the two of them, total AI-related R&D investment rose from $100 million in 2001 to $1.6 billion in 2019. Further, NSF’s AI-related funding rose from 1.1 percent to 14.5 percent of total allocated funding between 2001 and 2019. This spike in both volume and share of AI-related R&D from government agencies illustrates a long-term commitment of countries like the United States to direct public investment toward AI innovation. And while the NIH and NHS comprise only a part of total U.S. public R&D investment, other agencies must expectedly follow suit to ensure advancements in AI continue and encourage private-sector adoption of new AI technologies.
