Artificial intelligence research continues to grow as China overtakes US in AI journal citations

The artificial intelligence boom isn’t slowing yet, with new figures showing a 34.5 percent increase in the publication of AI research from 2019 to 2020. That’s a higher percentage growth than 2018 to 2019 when the volume of publications increased by 19.6 percent.

China continues to be a growing force in AI R&D, overtaking the US for overall journal citations in artificial intelligence research last year. The country already publishes more AI papers than any other country, but the United States still has more cited papers at AI conferences — one indicator of the novelty and significance of the underlying research.

These figures come from the fourth annual AI Index, a collection of statistics, benchmarks, and milestones meant to gauge global progress in artificial intelligence. The report is collated with the help of Stanford University, and you can read all 222 pages here.

In many ways, the report confirms trends identified in past years: the sheer volume of AI research is growing across a number of metrics, China continues to be increasingly influential, and investors are pumping yet more money into AI firms.

However, details reveal subtleties about the AI scene. For example, while private investment in AI increased 9.3 percent in 2020 (a higher increase than 2018 to 2019 of 5.7 percent), the number of newly funded companies receiving funds decreased for the third year in a row. There are several ways to interpret this, but it suggests that investors expect that the winner-takes-all dynamic that has defined the tech industry — in which digital economies of scale tend to reward a few dominant players — will be replicated in the AI world.

The report’s section on technical advances also confirms the major trends in AI capabilities, the biggest of which is the industrialization of computer vision. This field has seen incredible progress during the AI boom, with services like object and facial recognition now commonplace. Similarly, generative technologies, which can create video, images, and audio, continue to increase in quality and availability. As the report notes, this trend “promises to generate a tremendous range of downstream applications of AI for both socially useful and less useful purposes.” Useful applications include cheaper computer-generated media, while malicious outcomes include misinformation and AI revenge porn.

One area of AI research that seems like it’s just beginning to come into its own is biotech. The drug discovery and design sector received the most private investment of any sector in 2020 ($13.8 billion, 4.5 times more than in 2019), and experts canvassed for AI Index’s report cited DeepMind’s AlphaFold program, which uses machine learning to fold proteins, as one of the most significant breakthroughs in AI in 2020. (The other frequently cited breakthrough last year was OpenAI’s text-generation program GPT-3.)

One area where the Index AI report struggles to gauge progress, though, is in ethics. This is a wide-ranging area, spanning everything from the politics of facial recognition to algorithmic bias, and discussion of these topics is increasingly prominent. In 2020, stories like Google’s firing of researcher Timnit Gebru and IBM’s exit from the facial recognition business drove discussions of how AI technology should be applied. But while companies are happy paying lip service to ethical principles, the report notes that most of these “commitments” are non-binding and lack institutional frameworks. As has been noted in the past: AI ethics for many companies is simply a way to slow roll criticism.