NSF pushes for ‘large-scale’ AI research with new program
The National Science Foundation is taking the lead on large-scale, long-term federal artificial intelligence projects with plans to invest $200 million over the next six years.
On Tuesday the agency launched the National AI Research Institutes program, which expects to allocate $120 million in funding. Most of it will go to groups that already have a research plan. The goal is to create institutes within the NSF, led by collectives of universities, agencies, industry representatives and nonprofits. Some of the money also will go toward planning grants to researchers who are only just starting to formulate their ideas.
Plans that qualify for institute track will sign cooperative agreements for between $16 million and $20 million and four to five years — maximum $4 million a year — for research in one of six target areas. Awardees on the planning track will receive $500,000 grants for up to two years. The tracks will run concurrently.
“These institutes will accelerate the transition of AI innovations into many economic sectors while also nurturing and growing the next generation of AI researchers and practitioners,” said Erwin Gianchandani, an acting assistant director at NSF, in the announcement. The focus areas are:
• Trustworthy AI
• Foundations of machine learning
• Innovation in agriculture and the food system
• AI-augmented learning
• Accelerating molecular synthesis and manufacturing
• Discoveries in physics
Grant proposals are due January 28, 2020, for the institute track and January 30, 2020, for the planning track.
NSF’s program is a partnership with the National Institute of Food and Agriculture, Department of Homeland Security Science & Technology Directorate, Federal Highway Administration, and Department of Veterans Affairs.
The agency teamed with others earlier this year to update the National AI Research and Development Strategic Plan. AI advances are key to several of NSF’s “10 Big Ideas” for future investments.