Monday, 29 January 2024: 10:45 AM
345/346 (The Baltimore Convention Center)
We provide a short update to the broader AMS community on the AI developed by the NSF AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography (AI2ES). Our goal is to develop novel AI techniques that are demonstrated to be trustworthy and that will directly improve prediction, understanding, and communication of high-impact weather and climate hazards. Launched in September 2020, AI2ES is conducting transformative convergent research across AI, weather, water, and climate applications, and risk communication. In addition to developing novel AI techniques, we focus on the nature of trust and what makes AI trustworthy for professional and scientific end-users.
In this talk, we overview the highlights of our research achievements for year 3 as well as our achievements in broader impact and workforce development. These include foundational work in risk communication; in weather, climate and coastal oceanography; and in AI, including uncertainty quantification for AI models. Particular focus is on investigations of forecasters' trust in AI products, and on approaches to predicting tropical cyclones, coastal fog, ocean eddies, and convective hazards. We also report on our extensive outreach efforts to all levels, including K-20 and workforce development.
Our website, ai2es.org, contains more about the full scope of our research.

