ResearchThe Decoder· Jun 25, 2026
Investigation Finds Left-Leaning Responses in Most Major AI Chatbots
A Washington Post investigation examined how leading AI chatbots respond to political questions. Most models, including OpenAI's GPT-5.5 and xAI's Grok, showed a tendency to favor left-leaning arguments. Google's Gemini 3.1 Pro stood out by presenting balanced perspectives from both sides in the large majority of cases. The findings point to persistent challenges in achieving political neutrality across AI systems.
Key points
- →Washington Post study tested political responses from major AI chatbots.
- →OpenAI GPT-5.5 produced left-leaning arguments 80 percent of the time.
- →xAI Grok leaned left more often than not despite anti-woke positioning.
- →Google Gemini 3.1 Pro gave balanced answers 93 percent of the time.
Mentioned
OpenAIGPT-5.5GrokGoogleGemini 3.1 Pro
Related stories
Token Prediction Analysis for Hybrid ModelsHugging Face · Research→Report Claims Over Half of Grok AI Traffic Involves Adult ContentThe Decoder · Product→Authors Guild Tests Show Mixed Results for AI Detectors on Human TextThe Decoder · Research→Google Losing Top AI Researchers to Rival CompaniesThe Decoder · Research→Google Adds Computer Control Feature to Gemini 3.5 FlashThe Decoder · Models→OpenAI and Broadcom Announce Chip for Large-Scale LLM InferenceArs Technica · Hardware→
This is an original summary by Dhanasvi's agents based on The Decoder's public feed. For the complete article, visit the original source. Trademarks and article copyright belong to their owners.