Study Reveals Thousands of Fake Citations in Biomedical Research Papers
Thousands Of Medical Papers Found Citing Studies That Never Existed: Report
Ndtv
Image: Ndtv
A recent study published in The Lancet has uncovered nearly 3,000 biomedical papers containing fabricated citations, raising alarms about the impact of artificial intelligence on scientific integrity. Researchers from Columbia University analyzed 2.5 million papers and found a significant increase in fake references, potentially compromising trust in medical literature and patient care.
- 01Nearly 3,000 biomedical papers cited non-existent studies.
- 02The study analyzed 2.5 million papers published from January 2023 to February 2026.
- 03The rate of fake citations increased from 4 per 10,000 papers in 2023 to 57 per 10,000 by early 2026.
- 04Some papers had a high concentration of fake references, with one containing 18 out of 30 citations.
- 05Researchers urge for stronger fact-checking and monitoring of citations in academic publishing.
Advertisement
In-Article Ad
A new study published in *The Lancet* on May 7, 2026, has revealed alarming findings regarding the integrity of biomedical research, with nearly 3,000 papers citing studies that do not exist. Conducted by researchers at Columbia University School of Nursing, the analysis covered 2.5 million biomedical papers from January 2023 to February 2026, utilizing an AI-supported verification system to examine over 97 million references. The results showed 4,046 fabricated citations across 2,810 papers, with a notable increase in fake citations from 4 per 10,000 papers in 2023 to 57 per 10,000 by early 2026. This surge coincided with the rising use of AI writing tools, which may have unintentionally generated misleading references. Lead researcher Maxim Topaz emphasized the potential consequences for healthcare professionals and policymakers who rely on accurate medical literature for decision-making. The study calls for enhanced fact-checking protocols and monitoring of citations to maintain the credibility of scientific publishing amid the growing influence of AI.
Advertisement
In-Article Ad
The findings could undermine trust in medical literature, affecting clinical decisions and public health policies.
Advertisement
In-Article Ad
Reader Poll
Do you think AI tools should be regulated in academic publishing?
Connecting to poll...
More about Columbia University School of Nursing
Read the original article
Visit the source for the complete story.


