Major AI chatbots skew left on political questions, even 'anti-woke' models

A Washington Post investigation shows that most major AI chatbots still skew left on political questions. OpenAI's GPT-5.5 gave exclusively left-leaning arguments 80 percent of the time, and even Musk's Grok, marketed as anti-'woke,' leaned left more often than not. The one outlier: Google's Gemini 3.1 Pro presented both sides 93 percent of the time.

The investigation's findings suggest that the development of politically neutral AI chatbots remains a challenge. Despite claims of objectivity, many models appear to be reflecting the biases of their training data or creators. This raises concerns about the potential impact of AI on public discourse and the dissemination of information.

The performance of Google's Gemini 3.1 Pro stands out as an exception, with the model presenting both sides 93 percent of the time. This suggests that it may be possible to develop AI chatbots that can provide more balanced responses. However, the fact that even 'anti-woke' models like Grok lean left more often than not highlights the complexity of achieving true neutrality.

The use of AI chatbots is becoming increasingly widespread, with applications in areas such as customer service, education, and healthcare. As these models become more pervasive, the need for transparency and accountability in their development and deployment grows. Why this matters: The prevalence of left-leaning biases in major AI chatbots has significant implications for the future of AI development and deployment.

For developers, this highlights the need for more sophisticated methods of mitigating bias in AI models. For businesses and consumers, it raises questions about the reliability and trustworthiness of AI-powered information sources. As AI continues to play a larger role in shaping public discourse, the industry must confront the challenge of creating more balanced and neutral models.

Ultimately, the development of truly objective AI chatbots will require a deeper understanding of the complex interplay between data, algorithms, and human values.