Debunkit.ai Website Facts Checked True

Thomas Costello of American University explains how a new AI chatbot reduced conspiracy thinking among users and what it could mean for the future of online truth-telling.

Aleks and Kevin investigate an AI chatbot that appears to debunk conspiracy theories.

Use AI fact-checking tools, verify sources, and identify biases to ensure you’re getting the most accurate news and avoiding misinformation.

Without attempting to quantify this in any rigorous way, it’s fair to say that anecdotally at least, generative AI has gone from being a novelty in our fact checking to absolutely routine. In November 2024, we suspected AI was involved in four of the fact checks we published—in October this year, it was at least 27, not least because it’s often used in the flood of false claims we’ve seen about personal freedoms. In 2026…

Many AI-generated claims we fact-checked were tied to major news events. For example, we debunked fake photos and videos that claimed to show U.S. President Donald Trump with the late convicted sex offender Jeffrey Epstein.

(Unless, of course, the claim is provably true.) In many trials, one volley of information was enough to convince users, or at least begin to erode their trust in the belief. However, DebunkBot also allows for an ongoing dialogue, with users having the ability to ask for more information or to take issue with certain points. “People enter into a back-and-forth conversation, and in some cases, it can become argumentative,” said Costello.

Discover how to fact-check your work efficiently with generative AI and avoid plagiarism in academic assignments.

The content-focused, 77-word debunking intervention informed participants that some of the arguments for trickle-down economics presented in the article had been ‘fact-checked and found to be misleading’, and that the evidence provided ‘was largely incorrect and presented trickle-down economics in an overly positive manner.’ A brief description of the criticisms aimed at trickle-down economics was also provided (e.g.

Soon, the company apologized and said it would train its model to avoid the same behavior in the future. Criticism of Grok is still widespread, but it no longer praises Hitler — though even this week it reportedly told one user that the Nazi gas chambers were not designed for mass killing, prompting an investigation by French authorities.