NSF-CSIRO: HCC: Small: From Legislations to Action: Responsible AI for Climate Change

Climate change is a key threat in contemporary life. To curb and ultimately reverse climate change governments, organizations, and citizens around the world are taking actions to reduce the carbon footprint of human activity. Achieving these goals requires the effective creation and implementation of climate interventions, ranging from the individual to the national level. Yet developing effective climate change policies poses a significant challenge. Numerous policy options are possible, and policies interact with one another in unexpected ways. The task is further complicated by the fact that there are many sources of information about climate change, but they are often laden with jargon and may conflict with one another. The goal of this project is to develop responsible artificial intelligence (AI) to help stakeholders make sense of climate change related information, so that non-experts can collectively make sense of and implement effective policies. This encompasses a wide variety of possible AI-assisted tasks, ranging from summarizing the latest news reports to explaining the outcome of new legislation. In particular, we will focus on improving policymaking and public awareness by bridging the information gap between a wealth of publicly available information and a dearth of actionable insights. This will ultimately improve the effectiveness of climate change policies. The technical aims are structured along three thrusts: 1) understanding which tasks users are willing to delegate to AI and why; 2) developing a large-scale dataset of climate change related information, ranging from legislative texts from Australia and the US to news articles and social media; and 3) developing prototypes of AI systems to demonstrate the promise of responsible AI in promoting effective climate change policies. This is a joint project between U.S. and Australian researchers funded by the Collaboration Opportunities in Responsible and Equitable AI under the U.S. NSF and the Australian Commonwealth Scientific and Industrial Research Organisation (CSIRO). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.