Moderate evidence
Published recordHealthProblem Space
More
Evaluating Artificial Intelligence Chatbots for Patient Education on Prostate Biopsy in the Prostate Cancer Diagnostic Pathway
- Truvace record
- TRV-2026-2987raw data (JSON)
- Version
- v1Published
- First published
- Evidence type
- Benchmark or lab evaluation
- Sources
- 1 sourcePeer-reviewed
- sha256
- 620c9f7df415…check it
Prostate biopsy is an important step in the diagnosis of suspected prostate cancer. Many patients look for online information before the procedure, and artificial intelligence chatbots are becoming part of this process. This study evaluated the quality, readability, actionability, transparency, and guideline concordance of prostate biopsy information generated by ChatGPT, Gemini, Copilot, and Perplexity. A standardized English-language question was submitted once to the free version of each platform on March 18,…
The problem
In a test of ChatGPT, Gemini, Copilot and Perplexity, five urologists found every chatbot's prostate biopsy answer was written at an advanced reading level, and guideline concordance differed between platforms.
What this doesn’t fix
Each chatbot answered a single standardized English-language question, submitted once to its free version on one date, and only the first response was analyzed.
Sources
1 on file- Journal of Cancer Educationdoi:10.1007/s13187-026-03014-9
Cite this record
v1Truvace Impact Record TRV-2026-2987, v1: “Evaluating Artificial Intelligence Chatbots for Patient Education on Prostate Biopsy in the Prostate Cancer Diagnostic Pathway.” Truvace, 2026-11-17. /record/TRV-2026-2987 (accessed on the day you copy it). sha256 620c9f7df415134d…
Change history
1 versionEvery change made to this record since it was first published, and why. Nothing has changed since it went up.
Certified into the record
Check this record hasn’t been altered
How to check it yourself, without trusting this page
Download any version from /api/record/TRV-2026-2987 and hash its canonical field yourself, for example with shasum -a 256. You should get its content_hash. Each version’s text also ends with prev:and the hash of the version before it (version 1 points to 64 zeros). If anyone had changed a single character of any version after it was saved, those hashes wouldn’t line up.
ace