STEM Update #6, Close out, AI Institutes Round #4
Tuesday, August 8, 2023
Context: In my role as division director of IIS, I’m sending out a short message to the IIS mailing list on the Second Tuesday Every Month (STEM). Here’s the installment for August 2023.
I’m hit with a double whammy this month. In addition to the second Tuesday falling as early as possible, yesterday was the deadline for close out.
What’s “close out”, you say? Much more than I realized from the outside, NSF has a very structured yearly rhythm. There’s an annual budget and a strict date marking when one budget is over and another date marking when the next budget kicks in. At the end of a budget year, every dollar that was allocated in that year’s budget needs to be accounted for. It’s like an agency-wide $10B checkbook balancing act. (“Step right up folks, and see the death defying performance on the high wire!”) As you might imagine, things get kind of tense for a few weeks.
So, yesterday was the “drop dead” date (not a great metaphor to go with the high-wire description…) for DD concurs. In other words, every proposal that was under consideration as part of this budget year needed to have its fate decided and then made official by the division director (me) or deputy division director (Wendy) reviewing the relevant materials and signing off. Most proposals were decided a long time ago---program staff made and documented recommendations for them based on panelist feedback, Wendy or I did our DD concurring, and life marches on. But some proposals were close enough to the borderline that we decided to hold on to them for a bit to better see the overall content of the portfolio and also to find out whether money that was set aside for another purpose ended up becoming available. Some other proposals were held because they involve complex negotiations with other agencies or other units within NSF and it took time to hash everything out. And of course some panels are simply held late in the fiscal year. But close out is when all of that comes to a crescendo and a final determination needs to be made on everything. Operations staff are making urgent phone calls to other divisions, program staff is working through revised project budgets with their PIs, and Wendy and I are DD concurring like there’s no tomorrow. Returning to the circus imagery, the entire troupe comes into the Big Top for the rousing finale! Ta da!
Anyway, that’s my excuse for not sending this message earlier today.
One other item I wanted to mention is that we announced last week the fourth solicitation for the National AI Research Institutes. At the moment, NSF and its partners have funded 25 institutes over 3 rounds (in 2020, 2021, and 2023) with a total investment of about $500M. While the basic structure of the AI Institutes has remained pretty consistent through the rounds, each round has its own set of themes. This round, the themes are: AI for Astronomical Sciences, AI for Discovery in Materials Research, and Strengthening AI. I’m very proud of all the AI Institutes, which are doing amazing work for the field and for the nation. But I’m particularly excited about this announcement because it’s the first that’s gone out since I’ve been at NSF where I was involved in all the stages of development from brainstorming to writing to approving the finalized solicitation. I can’t wait to see the next phase as talented researchers from all over the U.S. propose fantastic projects in response. Kudos to the program staff and operations staff who worked hard to bring the pieces together to make this happen. I’ll particularly call out Jim Donlon (see STEM #5), who did an amazing job lining up partners and crafting descriptions and overseeing the whole complex process.
For official details, see: https://www.nsf.gov/pubs/2023/nsf23610/nsf23610.htm . Thanks!
Until next time!
-Michael