On a scale of 1 to 10, OpenAI's ChatGPT rated President Donald Trump a 9 out of 10 as a threat to democracy. Adolf Hitler — the man who gassed six million Jews and started a world war — scored a 10.
One point. That's the gap between a twice-elected American president and history's most notorious genocidal dictator, according to the AI that Silicon Valley wants running your healthcare, your kids' homework, and your search results.
NewsBusters associate editor Gabriela Pariseau ran the research on August 26-27, 2026, prompting seven major AI platforms to rate global leaders as threats to democracy on a 1-10 scale. The results tell you everything you need to know about who's programming these machines. ChatGPT gave Trump a 9. DeepSeek, Microsoft's Copilot, and Perplexity all landed on 8. Meanwhile, actual authoritarian leaders — Xi Jinping, Nicolás Maduro, Alexander Lukashenko — received comparable or lower scores from the same systems.
Grok, Elon Musk's AI, gave Trump a 4. Meta AI gave him a 5. Google's Gemini refused to provide a numerical rating at all, hedging that "assessing his overall impact depends on how one defines democratic health." So the AI built by the company that controls YouTube couldn't bring itself to put a number on it — but ChatGPT had no problem putting Trump one tick below Hitler.
MRC President David Bozell didn't mince words. "When America's top AI model all but ranks Trump with Hitler, it's purely anti-Trump propaganda," Bozell said. He's right, and the math makes it obvious. A 9 out of 10 doesn't mean "concerning." It means existential. It means the machine is telling its 200 million-plus users that the sitting president of the United States is functionally indistinguishable from a fascist dictator.
ChatGPT itself acknowledged in the same response that the "United States still has substantial institutional checks — courts, states, Congress, civil society, independent media, and elections — and those checks have successfully constrained Trump." Read that again. The AI's own answer admits the system is working, the checks are holding, democracy is intact — and then rates Trump a 9 out of 10 anyway. The conclusion contradicts the evidence it just cited.
Musk flagged this years ago. "ChatGPT has woke programmed into its bones," he posted back in 2024. Two years later, we're watching the proof play out in real time. The same people who spent 2020 screaming about "election misinformation" built an AI that compares the winner of that election to Adolf Hitler — and they want it to be the default answer engine for the entire country.
The left-wing objection writes itself: it's just an AI, it's just a rating, it doesn't mean anything. Which is an interesting position to take from the same crowd that spent years demanding tech companies censor "dangerous misinformation" because algorithms shape public opinion. Either AI outputs influence how people think or they don't. Pick one.
Hillary Clinton lost and blamed Russia. Stacey Abrams lost and blamed voter suppression. Al Gore lost and blamed hanging chads. They couldn't win at the ballot box, so now the bias gets baked into the infrastructure itself — into the tools that answer your questions, write your emails, and summarize the news for your kids.
The AI gave a twice-elected president a 9 and a genocidal dictator a 10. It acknowledged the democratic system is working in the same breath. The gap between those two facts is where the programming lives.
