AI models can write convincing fraudulent peer reviews that evade current detection tools, posing a new risk for research integrity.
Key Details
- 1Chinese researchers used the AI model Claude to review 20 actual cancer research manuscripts.
- 2The AI produced highly persuasive rejection letters and requests to cite irrelevant articles.
- 3AI-detection tools misidentified over 80% of the AI-generated reviews as human-written.
- 4Malicious use could enable unfair rejections and citation manipulation within academic publishing.
- 5AI could also help authors craft strong rebuttals to unfair reviews.
- 6Authors call for guidelines and oversight to preserve scientific integrity.
Why It Matters
Radiology and imaging AI research are heavily dependent on fair and trustworthy peer review processes. The inability to reliably detect AI-generated, fraudulent reviews threatens the credibility of published research and the overall progress of the field.
Source
EurekAlert
Related News
•EurekAlert
AI Predicts Risks for Outpatient Stem Cell Therapy in Myeloma
Researchers use machine learning to predict adverse events during stem cell therapy for multiple myeloma, improving outpatient safety.
•EurekAlert
AI-Enhanced CT Heart Fat Measurement Boosts Cardiovascular Risk Prediction
AI-derived measurement of heart fat from CT scans significantly improves long-term cardiovascular disease risk prediction.
•EurekAlert
Molecular Test BiliSeq Greatly Improves Bile Duct Cancer Detection
The BiliSeq molecular test developed at UPMC doubled detection sensitivity for bile duct cancer compared to standard pathology.