ChatNetZero: Harnessing Large-Language Models for Synthesizing Evidence to Evaluate Non-State Actors’ Climate Actions
H 0104 (Elinor Ostrom Hall)Large language models (LLMs) that power generative AI technologies like ChatGPT and Bard can analyze vasts amounts of textual data to generate easy to understand insights in human prose. But these technologies have also been demonstrated to hallucinate, generating convincing but false answers; prone to bias, and difficult to unpack due to the “black box” […]
Conference Theme: Climate Policy (Instrument) Evaluation
Research Methods: Digital evidence synthesis and machine-learning methods