In This Story
Training emergency communication operators and providing decision support to public officials is uniquely challenging. When someone calls 911, a human operator plays an essential role—gathering critical information, sensing urgency, and providing empathy to a person in crisis. Intelligent systems, however, have a growing function—supporting operators as they learn how to manage high-pressure situations and help officials make split-second decisions about how to guide the public during rapidly evolving emergencies. At these moments, human strengths and AI capabilities must work together.
Hemant Purohit, an associate professor in George Mason University’s Information Sciences and Technology Department, has built his career at this intersection. Supported by a grant from the National Institute of Standards &Technology (NIST) Purohit has worked with Fairfax County’s Department of Public Safety Communications to co-develop an AI-powered training system for 9-1-1 operators. This project combines expertise from George Mason’s Center for Resilient, Adaptable, and Safe Communities, the Humanitarian Informatics Lab, and the Sustainable Peace Lab, where researchers examine how to train people in high-stress, information-sparse environments of diverse emergency scenarios.
As director of the Humanitarian Informatics Lab, Purohit is on a mission to crack the interdisciplinary code that blends the science of design, analysis, and evaluation of information systems with the realities of organizational work in social welfare and governance. At the moment, one of his driving passions is empowering the human emergency infrastructures that serve as a community’s first line of response—helping them save time, reduce costs, and deploy resources more effectively at a moment when climate change and social instability make these needs especially urgent.
“A central question of my work is: How can we build systems that amplify the best of both human and AI teammates, especially when every second matters?” said Purohit.
Just this past fall, the National Science foundation awarded Purohit a $1.25 million grant to partner with the Virginia Beach Department of Information Technology and the Department of Emergency Management on PCExplorer (Physical & Citizen Sensing Exploration tool) — an AI-driven system designed to help officials guide a distressed public to safety through improved decision support for shelter planning and crisis communication. For this project, he is collaborating with a multidisciplinary team that includes Qian Hu from George Mason’s Schar School of Policy and Government; computer scientist Ayan Mukhopadhyay and infrastructure engineering expert Hiba Baroud of Vanderbilt University; and social scientists Joshua Behr, Rafael Diaz, and Wie Yusuf of Old Dominion University.
Purohit’s research sits at the sweet spot of information science and human-centered computing—a combination that’s rare but much needed for this kind of research. Some may be surprised at how much his research centers on understanding how humans navigate AI-mediated work environments and how systems can be designed to support—not supplant—human judgment. Building emergency training tools and decision-support systems requires deep insight into how operators interpret cues under pressure, how teams coordinate, and where AI can responsibly augment human expertise for decision-making.
“I carry this passion for science-driven tools for social good and strongly believe that the complex problems of our society require a multidisciplinary research approach such as the agenda of human-AI collaboration research!” said Purohit.
Purohit’s research spans mining social and web data to support emergency communications and disaster relief, developing interactive AI tools for smart communities, and understanding how not only governments and responders but communities themselves engage with crisis information. He also contributes to the design principles that will guide next-generation AI systems.
But his research is not just theoretical or empirical; it is producing real technologies that are being used by local communities and governments in response to natural disasters (e.g., hurricanes), violence, and online fraud. Across these efforts, Purohit works to ensure that as AI becomes a larger part of crisis preparedness and response, it does so in ways that strengthen—rather than replace—the people who keep communities safe.