The reduction in conspiracy belief caused by DebunkBot conversations remained just as large two months later as it was immediately after the conversation.
Notes on verification
Original Science paper (Costello, Pennycook, Rand 2024) and multiple independent news/university sources confirm the durability of the effect at two-month follow-up. However, a 2026 Editorial Expression of Concern flagged data-merging errors and reporting inconsistencies in the underlying dataset; authors claim a corrected analysis reproduces the original findings in direction, significance, and size, but full independent re-verification is pending, warranting some caution. [tier=gold indep_score=0.87 clusters=5 claim_tier=notable]
Sources
- Chatbots are surprisingly effective at debunking conspiracy theories (seed:technology_and_ai)
- https://www.science.org/doi/10.1126/science.adq1814 (corroboration)
- https://mitsloan.mit.edu/ideas-made-to-matter/mit-study-ai-chatbot-can-reduce-belief-conspiracy-theories (corroboration)
- https://retractionwatch.com/2026/06/11/science-ai-chatbots-debunk-conspiracy-theories-expression-of-concern/ (corroboration)
- https://undark.org/2024/10/30/could-ai-help-curb-conspiracy-theory-beliefs/ (corroboration)
- https://boingboing.net/2024/11/01/send-the-conspiracy-theorists-in-your-life-to-the-debunk-bot.html (corroboration)
- https://www.science.org/content/page/2026-aaas-newcomb-cleveland-prize-winners (corroboration)