Political Biases in LLMs: Literature Review & Current Uses of AI in Elections
TL;DR: This research discusses political biases in Large Language Models (LLMs) and their implications, exploring current research findings and methodologies. Our research summarized eight recent research papers, discussing methodologies for bias detection, including causal structures and reinforced calibration, while also highlighting real-world instances of AI-generated disinformation in elections, such as...
Mar 7, 20246