cuts to – A small NSF directorate that funds social, behavioral and economic research is facing steep pressure in the Trump administration—sparking warnings that the consequences could land far beyond academia, from earthquake early warnings to the safety of future arti
For earthquake warnings to work, the science has to do more than predict shaking. It has to get people to sign up, receive alerts, and understand exactly what to do when the message arrives.
Lori Peek. a sociologist at the University of Colorado Boulder who studies disaster mitigation. says that kind of “getting predictions to people in a meaningful way” has depended on federal research into how people think and behave. Peek has described her work building an early warning system for earthquakes in the U.S. in 2019—and she argues that behavioral science is what turns an alert into an action.
That kind of research is funded in part by a small NSF directorate called the Directorate for Social, Behavioral and Economic Sciences, or SBE. It makes up a tiny fraction of the federal budget, but researchers say it has produced outsized returns.
The money for this research has not been technically rescinded, but the funding pipeline has already started to shrink. According to the science advocacy group Grant Witness. only seven SBE grants had been awarded as of May 28. while the federal government typically makes hundreds of such awards per year.
The pushback goes beyond administrative funding levels. Jon Freeman, a psychologist at Columbia University, says the field has been singled out as a target by the administration.
Freeman points to a year earlier. when Republican Senator Ted Cruz wrote a report that specifically called out many SBE-funded programs as. quote. “neo-Marxist class warfare propaganda.” He also points to President Trump’s 2027 budget proposal. which would eliminate the directorate and zero out funding for this kind of research.
In response to a request for comment, White House spokesman Kush Desai sent an email statement. It said, quote, “The Trump administration is committed to cementing America’s dominance in cutting-edge technologies of the future, not in ideologically driven social sciences,” unquote.
Advocates for the research argue the administration’s framing misses how deeply behavioral science is woven into technologies people use—and into crises people face. For computer scientist Aron Culotta at Tulane University. the stakes show up in the ways AI can fail. especially where humans and machines meet.
“If we think about the failure modes of AI and kind of what can go wrong, they’re almost all at that human-AI intersection point,” Culotta said.
Culotta says the technology behind systems like ChatGPT comes from computer science. but making them safe depends heavily on psychology and social science research. He points to areas such as mental health and the ways people can become attached to these systems. as well as concerns like deepfakes and whether people can trust the output of AI.
Tom Griffiths. a cognition and artificial intelligence researcher at Princeton. describes the work from a longer view—how government-funded science often pays off only after years. sometimes decades. His focus as a cognitive scientist is not just what works now, but what technologies may matter later.
“In the 1950s, researchers began developing something called artificial neural networks—computer systems loosely inspired by the human brain—that now power many AI tools,” Griffiths said. “That’s the advantage, he says, of the government investing in science. It doesn’t have to pay off for decades.”
He warns that if the government “takes our foot off the accelerator now,” the gap may not be obvious immediately. “There’s going to be a point in the future where suddenly there’s going to be a gap with what’s going on in other countries,” Griffiths said, “and we say, what happened?”
The concern, as he tells it, is that the answer could be traced back to “this moment when the government stopped investing in behavioral science.”
behavioral science NSF SBE directorate earthquake early warning disaster mitigation artificial intelligence safety ChatGPT deepfakes mental health grant funding Ted Cruz report Kush Desai statement
So they’re cutting science again… cool.
Wait this is the NSF stuff about earthquakes, right? But like if they cut “behavior” research how does that stop earthquakes? Seems like they’re just trying to make people panic less or something.
I don’t buy the whole “behavioral science turns alerts into action” thing. People get the alert and still ignore it, that’s on them. Also the article says only 7 grants but I’m guessing that’s normal? Idk seems like sensational clickbait.
This is wild because it sounds like they’re cutting funding for the part that tells people what to do during disasters, and then acting surprised when nothing changes. And the AI part?? Like we’re gonna lose safety tech because of politics? I feel like it’s always “small directorate” until it affects everyone.