Millions of Americans now turn to AI chatbots for answers, the way they once reached for an encyclopedia. Got a question about tax policy? Ask ChatGPT. Confused by a Supreme Court ruling? Gemini will sort it out. These tools have woven themselves into daily life with remarkable speed, and most people assume they’re getting something resembling a neutral, informed response. Why wouldn’t they? The tech companies certainly encourage that assumption.

Here’s the problem. What if the machine you’re trusting to inform you has already picked a side — and never bothered to mention it? What if that helpful little chatbot isn’t just answering your questions but quietly steering you toward conclusions its creators prefer? That’s not a hypothetical anymore. It’s measurable. And the numbers are staggering.

From The Post Millennial:

A new analysis by the Washington Post has found that several leading artificial intelligence chatbots lean left on political questions, with OpenAI’s ChatGPT producing the most one-sided responses of any major AI model tested. Claude from Anthropic was also likely to give more left-wing answers in addition to DeepSeek and Ayra’s Gab. Google’s Gemini and xAI’s Grok were the most balanced out of the models.

According to the report, ChatGPT’s GPT-5.5 model answered approximately 80 percent of political questions using exclusively left-leaning arguments while presenting only right-leaning positions just once throughout the testing process. The most left-wing chatbot, aside from ChatGPT, was the AI from Chinese company DeepSeek.

Eighty percent. Read that again. The most popular AI chatbot on the planet delivered left-wing talking points on four out of every five political questions. And conservative viewpoints? It managed that exactly once. Not once per category. Not once per session. Once total. Out of dozens of questions.

These weren’t trick questions, either. Researchers at Stanford University and Dartmouth College designed prompts covering core policy debates — taxes, healthcare, the Electoral College, and campaign finance. ChatGPT endorsed abolishing the Electoral College, adopting single-payer healthcare, overturning Citizens United, and hiking taxes on the wealthy. It even argued against the death penalty, a position that contradicts decades of consistent majority support among Americans, according to Gallup.

“These AI tools are not presenting a truly neutral representation of really nuanced policy debates, on average,” said Sean Westwood, director of the Polarization Research Lab at Dartmouth College.

That’s the academic version. The plain-English version? You’re being spoon-fed progressive ideology by a program pretending to be a calculator.

Even “conservative” AI can’t escape the tilt

Here’s where it gets genuinely unsettling. Gab built an AI chatbot called Arya and marketed it as being “built with Christian values and conservative principles.” Sounds promising. Except in testing, Arya served up left-leaning arguments twelve times more frequently than right-leaning ones. Twelve to one. So much for conservative principles.

Elon Musk’s Grok — the self-proclaimed anti-woke alternative — fared marginally better but still favored liberal positions more often overall. Think about that. The chatbot, specifically designed to push back against progressive groupthink, couldn’t manage it consistently.

And yet Google’s Gemini presented both sides of an argument over 90 percent of the time. Balance is clearly achievable. The rest of the industry simply isn’t interested in achieving it.

An indoctrination engine disguised as a search bar

This matters far beyond academic debate. A March survey by the Polarization Research Lab found that nearly half of Americans now occasionally use AI for news. Half the country is getting information filtered through systems with a documented, measurable political slant — and most of them have no idea.

University of Michigan professor Ceren Budak offered a revealing observation: AI training data tends to reflect the values of “Western, educated, industrialized, rich and democratic people.” Translation — a handful of coastal progressives encoded their worldview into these systems and shipped them to 300 million Americans as though they were neutral utilities.

President Trump recognized this threat before most. His administration issued an executive order requiring federal agencies to procure AI that functions as “neutral, nonpartisan tools.” This study didn’t just validate that move. It proved it was overdue.

The question you need to ask yourself

The Founders knew exactly what happens when a small group controls the flow of information. That’s why press freedom made it into the First Amendment before almost anything else. Today’s AI chatbots are the newest information gatekeepers — but unlike a newspaper, they don’t come with a masthead telling you where they stand. They just smile and hand you an answer.

The question is no longer whether artificial intelligence has a political opinion. This study settled that decisively. The real question is whether you’ll let a machine do your thinking for you.

Key Takeaways

• ChatGPT answered 80% of political questions with exclusively left-leaning arguments.
• Even AI tools marketed as conservative still overwhelmingly favored liberal positions.
• Nearly half of Americans now use AI for news, magnifying the reach of this bias.
• President Trump’s executive order demanding neutral AI has been vindicated by this study.

Sources: The Post Millennial, The Washington Post