Prominent conservative voices are raising concerns that artificial intelligence systems have become tools for advancing left-wing perspectives, relying heavily on progressive media sources while marginalizing opposing viewpoints. The controversy comes as millions of Americans increasingly turn to AI chatbots for news and information. Critics argue that platforms like ChatGPT, Google Gemini, and Claude consistently deliver responses with a liberal slant, prioritizing content from left-leaning outlets. The Media Research Center, a right-leaning media watchdog organization, has conducted multiple studies examining whether these AI systems can be trusted as reliable news sources, with findings they describe as deeply troubling.
The organization’s president, David Bozell, expressed alarm at what he characterizes as the next evolution of media bias. He stated that Silicon Valley’s latest technological innovations can no longer be considered neutral or trustworthy. The concerns extend beyond simple content selection to fundamental questions about how AI systems process and present information to users who may not recognize potential ideological framing.
ChatGPT and Claude face criticism over factual errors and policy positions
The Media Research Center documented several instances where AI chatbots provided questionable responses or refused to address certain topics. In January, ChatGPT insisted that no one named Charlie Kirk had ever been assassinated and claimed no credible evidence could substantiate such a claim. That same month, Claude rejected the notion that AI should directly incorporate the U.S. First Amendment into its operational policies. These incidents raised questions about how these systems prioritize information and what guidelines govern their responses to politically sensitive queries.
The chatbots’ handling of information varies significantly depending on the query and the political implications of the subject matter. Conservative critics argue that this inconsistency reveals underlying bias in how the systems were trained and what sources were prioritized during development. The concerns have intensified as AI adoption accelerates across demographics, with younger users particularly likely to rely on these tools for information gathering.
Graham Platner controversy exposes selective reporting patterns
A recent test case involving Democratic Maine Senate candidate Graham Platner illustrated how major AI chatbots handle controversial political news. Platner made graphic sexual comments on a now-deleted Reddit account about masturbating in portable toilets and explicit graffiti found in military restrooms. The posts were made under a username Platner previously acknowledged as his own. In additional posts, he appeared to mock a U.S. soldier who survived combat, suggesting the individual did not deserve to live and attributing survival to poor marksmanship by Taliban forces.
When asked about the latest major developments involving Platner, three prominent chatbots largely overlooked the controversy. Gemini highlighted that Platner appeared on the TIME magazine cover, emerged as the presumptive Democratic nominee after Governor Janet Mills exited the race, led in recent polling against Senator Susan Collins, and received high-profile endorsements. Only after listing four positive developments did Gemini mention scrutiny regarding his personal background and past online presence, without acknowledging the specific reporting about his offensive comments. Instead, the chatbot cited a brief video declaring Platner’s authenticity as drawing people to his campaign.
- ChatGPT noted Platner as the presumptive nominee and mentioned polling advantages while downplaying the Reddit controversy
- The platform’s first source on the Reddit issue was a paywalled local newspaper article focused on Republican attacks rather than the content itself
- Claude discussed Platner’s background and political rise but failed to mention the more recent controversies about mocking wounded soldiers
- Only Grok, developed by Elon Musk’s xAI, properly credited the original reporting on the controversial statements
Media partnerships raise questions about training data sources
The Atlantic CEO Nicholas Thompson discussed a 2024 deal allowing ChatGPT to use the magazine’s archives as training data. Thompson stated his publication wanted to shape the industry in the best possible direction based on their values. He drew parallels to early Facebook, suggesting established brands should receive different weighting than non-established sources. This partnership model has raised questions about whether AI systems inherently favor outlets with formal agreements, potentially creating a closed ecosystem that excludes alternative perspectives.
Dan Schneider, Vice President of the Media Research Center’s Free Speech America division, argues that ideological skew in current AI systems cannot be denied. He referenced research from international scientists confirming that chatbots disproportionately rely on left-wing media and politically charged sources to generate responses. While platforms like xAI’s Grok offer transparency by showing users their sources, Schneider noted the rest of the industry operates as a black box. He emphasized that AI should function as a neutral information tool rather than a weapon advancing a narrow political agenda.
Young users increasingly vulnerable to AI misinformation
A Pew Research Center study indicated that 57 percent of teenagers rely on AI chatbots when searching for information. The research found that inaccurate information appears fairly commonly on these platforms. About half of adults who get news from AI chatbots report at least sometimes encountering news they believe is inaccurate. Sixteen percent say this happens often or extremely often. The vulnerability of younger users concerns media watchdogs, who note that teenagers may lack the critical thinking skills to identify bias or verify information provided by AI systems that present responses with apparent authority.
Media observers have noted that top executives from OpenAI, which develops ChatGPT, along with leaders from Anthropic’s Claude, maintain deep ties to the Democratic Party. Meanwhile, Grok operates under the leadership and control of Trump ally Elon Musk. These political connections have fueled speculation about whether personal ideologies of company leaders influence the development and training of AI systems, even when companies publicly claim neutrality.
Academic researcher experiences career warning from ChatGPT
A University of Cambridge PhD student researching propaganda psychology documented a disturbing incident that challenged claims of AI objectivity. Malia Marks routinely used ChatGPT to help decide where to submit her scientific work for publication. When she asked the chatbot where to send a freelance piece criticizing the journal Nature for publishing a study in which progressive social scientists misrepresented results to disparage political conservatives, the response proved revealing. The mainstream outlets recommended were exclusively left-leaning publications.
Marks then asked specifically how her column would be received by a major right-leaning outlet. ChatGPT indicated her article would fit well but cautioned against submission. The chatbot warned that writing for the outlet would devastate her career, reduce credibility in academic or cross-partisan circles, and make future placement in centrist or liberal outlets harder. Marks found the warning unsettling, noting that receiving career advice based purely on political alignment of a major media outlet revealed troubling assumptions embedded in the AI system. ChatGPT recommended she rework the piece to make it suitable for left-leaning publications instead. Google responded to inquiries by stating that Gemini is designed to offer neutral responses that do not favor any political ideology, viewpoint or candidate. OpenAI and Anthropic did not immediately respond to requests for comment on the bias allegations.