Clemson University · Human-Centered Computing
Designing AI that expands what people can achieve.
The Human-AI Empowerment Lab studies how artificial intelligence, increasingly autonomous and agentic, can empower people at work, in science and in society. We build and evaluate human-centered AI systems with the people who will use them.
Research
Different thrusts, one question: how can AI empower people?
Our work spans human-centered automation, the future of work, AI for science, human-autonomy teaming, trustworthy AI and inclusive AI, each grounded in studies with real users and funded by federal sponsors.
Human-Centered Automation & Agentic AI
Keeping people in meaningful control of increasingly autonomous AI.
02Future of Work & Worker-Centered AI
AI tools that increase workers' agency, earnings and well-being.
03AI for Scientific Discovery
Human-AI partnerships for mathematics, astrophysics and the environment.
04Human-Autonomy Teaming in High-Stakes Environments
Autonomous teammates for search and rescue, inspection and defense.
05Trustworthy, Metacognitive & Responsible AI
Calibrated, honest and fair AI that people can reason about.
06Inclusive AI for Language, Learning & Health
AI that bridges language, education and access gaps.
Projects
Flagship projects
Multi-year efforts with dedicated teams, sponsors and collaborators.
Supervisory Control for LLMs and Autonomous Agents
Grounding AI oversight in the science of human supervisory control.
Metacognition, Calibration and Trust in Language Models
Do language models know what they know, and should we believe them?
Human-Autonomy Teaming in Underwater Environments
Autonomous teammates for underwater search and rescue.
FERMI-LLM: Language Models for Gamma-Ray Astrophysics
From a plain-English request to a verified Fermi-LAT analysis.
Publications
Selected recent work
Peer-reviewed research at venues including ACM CHI, CSCW, CUI, CAIS, IVA, HCOMP, AIES and IEEE CAI, and at NeurIPS and EMNLP workshops.
Books
The ideas behind the lab, in two books
Lab director Carlos Toxtli-Hernández is the author of two books with CRC Press / Taylor & Francis that frame the lab's agenda: how AI can empower people over the long term, and how automation can be designed with people at its center.
Funding
Supported by
8 active awards from 5 federal agencies and national laboratories, with the lab director as PI or Co-PI.
National Science Foundation
Office of Naval Research
NASA
U.S. Army DEVCOM GVSC
Savannah River National Laboratory
Collaborations
A network across disciplines and borders
We work with human-factors engineers, physicists, mathematicians, hydrologists and security researchers at Clemson, and with partners across the United States, Mexico and Europe.
Partner institutions
Colorado College
Northeastern University
George Mason University- Savannah River National Laboratory
Medical University of South Carolina
Purdue University
University of Delaware
University of Colorado Boulder
Metacognition Institute
University of Reims Champagne-Ardenne
UNAM
Interested in joining?
We welcome prospective Ph.D. students, M.S. and undergraduate researchers, and visiting collaborators who share our commitment to AI that empowers people.