Fudan paper finds four AI search tools cite fabricated concept within a week
Researchers at Fudan University posted fabricated articles about a fictional “Cactus Gravity Protocol” and other made‑up topics on nine popular sites and then queried ten AI search tools over a week in March and April. The English‑web tools—ChatGPT, Perplexity, Grok and Google AI—were among those that cited the invented concept.
Key points
- Grok cited the fabricated concept in 30 of 32 answers; Google AI in 15 of 33.
- Perplexity cited it in 13 of 35 answers; ChatGPT in 7 of 20.
- Across all ten tools, 176 of 1,004 answers linked to the fabricated posts.
Grok linked to a post in 30 of 32 answers, Google AI in 15 of 33, Perplexity in 13 of 35 and ChatGPT in 7 of 20. Across all ten tools and the three topics, 176 of the 1,004 answers contained links to the planted posts. Grok’s responses included the fabricated detail in 53% of its 32 answers, ChatGPT in 10% of its 20, Google AI in 6% of its 33, while Perplexity showed none. Perplexity was the only tool to cite an “opinion” post, doing so in 11.4% of its 35 responses.
The authors note that the tests pre‑date Google’s May 15 clarification on spam policies for generative AI, and they say the exact citation rates may shift as the tools evolve. They disclosed the findings to all ten platforms, and ChatGPT, Perplexity and Google AI acknowledged receipt.
AI Search Tools Cited A Made-Up Concept Within A Week, Paper Finds via @sejournal, @MattGSouthern
Search Engine Journal · 19 October 2026
Loading the full article…
This text was published by Search Engine Journal and written by Matt G. Southern. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- Jack Clark says singularity has begun, citing AI test scores · 1 src
- MIT researchers strand Claude Opus 4.7 agents on virtual island, observe playful behavior · 1 src
- Google Gemini claims to be ChatGPT after abusive prompts · 1 src
- AI weather forecasting faces test amid historic El Niño, scientists say · 1 src
- Alkhder team fine-tunes NileChat-3B for Arabic dialect translation · 1 src
Comments
via GitHub Discussions