Angel Hsu
University of North Carolina at Chapel Hill, NC·Earth & Environmental Science
Companies increasingly make public promises to cut greenhouse gas emissions, and people now often learn about those promises through generative artificial intelligence (genAI), tools such as chatbots that write text in response to questions. If these tools give wrong or misleading answers about who is doing what on climate, they could weaken public trust, distort policy support and change how people behave.
- Fields
Earth & Environmental Science
Social Science
Computer Science
- Topics
generative AI
genAI
artificial intelligence
- Grants
- 1 known grant, through Dec 2026
What they work on
- Systematically test generative AI models, such as chatbots, to find out how accurately they report information about private-sector climate commitments.
- Design new metrics and benchmarks that score the accuracy and credibility of AI-generated information about corporate climate pledges.
- Convene interdisciplinary pilot projects that examine how generative AI affects corporate climate behavior, governance, public trust and policy support.
- Carry out a stakeholder gap analysis and policy framework review to set the aims and flagship pilot projects of the planned CLAIM center.
More about the research
This planning project aims to lay the groundwork for a research center that studies how genAI shapes climate information and how it could instead help track and verify climate action.
Why it matters
Regulators, researchers and the public need reliable ways to judge whether corporate climate pledges are real. By building tests for the accuracy of AI-generated climate information and releasing open datasets on different actors' climate impacts, this work could help regulators and citizens tell credible sources from unreliable ones. It could also make it harder for companies to overstate their climate progress.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Bo Luo
University of Kansas, KS·Computer Science
Define the key scientific questions on how generative AI threatens and could strengthen security and privacy, such as privacy harms from training data and synthetic data for attack detection.
Rui Zhang
Pennsylvania State University, PA·Computer Science·Early-career
Develop controllable summarization models that satisfy fine-grained, compositional user preferences for both the content and the style of a summary, along with rigorous task definitions and evaluation metrics.
Gregory Nelson
University of Maine, ME·Education Research·Early-career
Use design-based research to iteratively develop and evaluate peer reflection groups that build professional skills including iterative improvement, lifelong learning, GenAI use and psychological safety, with minimal instructor time.
The school
University of North Carolina at Chapel Hill: 72 labs on PI Pages, $1.6B a year on research, 523 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.