Model13h ago
AI Chatbots Maintain Left-Leaning Political Bias
OpenAI's GPT-5.5 provided exclusively left-leaning arguments 80 percent of the time in a recent investigation. Even Grok skewed left despite its anti-woke marketing. Only Google's Gemini 3.1 Pro remained neutral, presenting both sides 93 percent of the time. These results suggest that systemic training biases persist across diverse model architectures and corporate philosophies.