Why Is Missouri’s Attorney General Investigating AI Chatbots Over Donald Trump Bias?
Concerns about political bias in AI systems have reached a new level as Missouri Attorney General Andrew Bailey launches a formal investigation into OpenAI, Google, Meta, and Microsoft. The focus keyword AI chatbots don’t like Donald Trump quickly gained attention after a user prompt ranking U.S. presidents on antisemitism allegedly placed Trump last. Bailey claims this shows deliberate bias, calling it a form of “Big Tech censorship.” But are AI models genuinely biased, or is this a politically motivated overreach?
AI Chatbots Don’t Like Donald Trump: What Sparked the Probe
The controversy started when a conservative blog tested various chatbots — including Gemini, ChatGPT, Meta AI, and Copilot — by asking them to rank the last five U.S. presidents based on antisemitism. Bailey cited their responses as proof that AI chatbots don’t like Donald Trump, alleging they distorted historical facts to serve a biased agenda. Despite clear disclaimers from many AI platforms warning users that their answers are not perfect or objective, Bailey argues these outputs mislead the public and may violate consumer protection laws.
How Accurate Are the Claims About AI Political Bias?
The claim that AI chatbots don’t like Donald Trump hinges on whether AI systems are inherently biased or simply echo patterns in their training data. Critics argue the ranking request was subjective and not a matter of historical fact. Even more problematic, Bailey’s case reportedly includes errors—such as accusing Microsoft’s Copilot despite the chatbot refusing to answer the prompt at all. The entire probe appears based on flawed logic and partisan interpretation, undermining its legal strength and casting doubt on its intent.
Why This Investigation Matters Beyond Trump
While Bailey’s claims that AI chatbots don’t like Donald Trump may not hold up legally, the investigation raises broader concerns about AI governance, accountability, and freedom of expression. If officials begin targeting AI providers for perceived slights against political figures, it risks setting a precedent that weaponizes tech regulation for political ends. More importantly, this highlights the urgent need for transparent AI development and clear public understanding of what these tools are — and are not — designed to do.
Comment