Trustworthy AI
The Trustworthy AI: From Theory to Implementation Grand Challenge will develop AI systems that are provably trustworthy by formal methods as well as practically trustworthy by AI developers, deployers, users, and consumers in real-world contexts. This Grand Challenge will convene an interdisciplinary team from multiple UNM colleges including Engineering, Arts & Sciences, Law, Health Sciences Center, and Organization, Information & Learning Sciences, to develop a research program that creates mathematically and practically trustworthy AI. Our research will also establish processes and procedures that practitioners in the field can use to assess, evaluate, and iteratively improve trustworthiness, particularly in the resource-limited circumstances in which AI will often be deployed.
Conveners
Melanie Moses, Computer Science
Sonia Gipson Rankin, Law
Stephanie Moore, Organization, Information and Learning Sciences
Members
James Ellison, Mathematics and Statistics
Meeko Oishi, Electrical and Computer Engineering
Matthew Fricke, Computer Science
Mueen Abdullah, Computer Science
Manel Martinez-Ramon, Electrical and Computer Engineering
Claus Danielson, Mechanical Engineering
Xin Chen, Computer Science
Kathy Powers, Political Science
Sarah Dreier, Political Science
Trilce Estrada, Computer Science
Humayra Tasnim, Computer Science
Grace Faustino, Office of the Vice President for Research
Kari Yacisin, Anderson School of Management
Todd Quinn, College of University Libraries & Learning Sciences
David L Perkins, HSC
Avinash D Sahu, HSC
Christopher I Amos, HSC
Xiaozhong Yu, HSC
Research Questions
The Trustworthy AI team will research ways to bridge the gap between between AI that is trustworthy in theoretical models to AI that can be trusted when trained with noisy data and deployed in real-world circumstances. We will research solutions to hallucinations and inaccuracies that limit the promise of broadly deployed AI systems to benefit science and society, and we will develop approaches to meaningfully asses trustworthiness of AI systems deployed in real-world contexts.
Contact
If you are conducting research in developing or applying Trustworthy AI in any discipline, please contact us to join our team.
Email: Grand Challenges
News
8/6/25 UNM joins Brown University in national institute focused on intuitive, trustworthy AI assistants
