France is going big on artificial intelligence (AI). President Emmanuel Macron yesterday announced a €1.5 billion plan to turn his country into a world leader for AI research and innovation, a field dominated by the United States and China. It calls for a hefty investment, a handful of specialized institutes, a focus on ethics and open data, and a call to recruit foreign researchers and French scientists working abroad to the country, not unlike Macron’s 2017 “Make Our Planet Great Again” climate initiative.
Macron presented his plans in a lengthy speech peppered with erudite references and touches of humor at the end of the“AI for Humanity” conference in Paris. Turning the country into an AI leader would allow France to use AI for the public good and ensure that a “Promethean” promise doesn’t become a “dystopia,” he said.
“The timing couldn’t be more perfect,” says Alessandro Curioni, vice president of Europe and director of IBM Research in Zurich, Switzerland, who attended the event in Paris. “We are becoming awash in data and we simply can’t keep up. The only answer is AI,” he says.
The government's measures are largely based on recommendations laid out in a report, presented yesterday by France’s star mathematician and Fields Medal winner Cédric Villani, a member of Parliament for Macron’s political group. Macron didn’t pick up on Villani’s suggestion to double the salaries of early career scientists to make academic research jobs more attractive, but a good part of the president’s plan aims at curbing the hemorrhage of talented researchers leaving France.
“There is a global arms race going on and the price to compete keeps going up and up,” says Jonathan Schaeffer, a computing scientist at the University of Alberta in Edmonton, Canada, adding that France should review progress annually and adjust funding levels if necessary. “It is hard to envision the future, but with the potential that AI has, a major investment today might be insufficient tomorrow,” he says.
The plan includes a national research program to be led by the French National Institute for Computer Science and Applied Mathematics (INRIA) in Rocquencourt and centered around four or five “dedicated institutes” anchored in existing research sites across the country.
The timing couldn’t be more perfect. We are becoming awash in data and we simply can’t keep up. The only answer is AI.
All in all, the government will spend €1.5 billion on Macron’s initiative, including €400 million on competitive calls for proposals, as well as €100 million “in the coming months” to help launch startup companies in hopes of attracting another €500 million from private companies, Macron said. But few details are available at this stage. “The scale will depend on the means that will be allocated” to the program, says François Sillion, INRIA’s interim CEO, adding that things will become clearer in the coming weeks.
Recruiting researchers will be the biggest challenge, says Karteek Alahari, a researcher at INRIA in Grenoble. France is a better place to live and work than some scientists think, says Alahari, who studied in India and did his Ph.D. and postdoc in the United Kingdom. He says the quality of students, freedom to pick research problems, and the quality of academic research attracted him to France. To match the flexibility of universities in other countries, Macron said he would seek to raise the time that publicly funded scientists are allowed to spend working in private companies, from 20% to 50%.
France is not the only European country that has grand ambitions in AI. Germany, too, wants to be a pioneer, Anja Karliczek, Germany’s minister for education and research, said at the conference. Both countries could link up their data centers and start bilateral research programs, she said, while Macron said part of the funds would be earmarked for Franco-German research projects. European Research, Science, and Innovation Commissioner Carlos Moedas also praised Macron’s vision. The European Union is set to unveil its own AI strategy on 25 April.
Macron’s strategy includes a focus on AI ethics to ensure that algorithms are controlled and work for the greater good, avoiding the “opaque privatization of AI or its potentially despotic usage” by foreign governments, the president said. He even proposed to set up an “IPCC of AI,” a group akin to the Intergovernmental Panel on Climate Change, which is based in Geneva, Switzerland, and made up of thousands of volunteer scientists who review climate science literature.
Several companies, including Microsoft and Fujitsu, timed announcements about new investments in France to Macron’s speech. DeepMind, a London-based leading AI company bought by Google in 2014, said it would open a Paris center this summer for research on AI, machine learning, deep learning, and reinforcement learning. “I think it’s an excellent choice,” said Rémi Munos, a DeepMind scientist currently working in London who will lead the Paris lab, in a video statement. “Not only because it allows me to go back to France, but also because Paris is in the process of becoming a major location for research in AI and machine learning.”