ESG Discourse and AI Semantic Analysis in Urban Digital Twin Enterprises: A Comparative Study of China and the United States

This code explores how large language models (LLMs), specifically Google Gemini, can be prompted to perform semantic analysis on corporate sustainability reports. Focusing on two urban digital twin enterprises from China and the United States, respectively, you will design and apply a prompt that guides the LLM to assess report content across specific ESG (environmental, social and governance) criteria and regulatory/cultural contexts. Through a comparison with a traditional quantitative approach, you will evaluate Gemini's ability to not only compare the reports' content, but also insights into the discourse, emphasis, and narrative framing within the sustainability reports. This lab combines natural language processing, prompt engineering, cultural analysis, and critical reasoning to examine the strengths and weaknesses of LLMs in AI semantic analysis.