A new research paper details how Large Language Models (LLMs) can be used to analyze over 150 years of German parliamentary debates on migration. The study found that while LLMs like GPT-5 and gpt-oss-120B can achieve annotation accuracy comparable to human agreement, their systematic errors require statistical correction for reliable long-term trend analysis. The research traced a shift from post-war solidarity towards migrants to increased anti-solidarity in the last decade, particularly since 2015. AI
IMPACT Demonstrates LLMs' capability for large-scale social science research, while highlighting the need for rigorous validation and statistical correction of their outputs.
RANK_REASON Academic paper detailing methodology and findings using LLMs for social science research. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →