‘Research could be sabotaged by flood of low-quality clutter’

Security researcher Tiberius Ignat warned at the Earma annual conference that the research system faces a new threat of sabotage by flooding it with low-quality applications and outputs. This sabotage, he said, could be driven by artificial intelligence. The warning identifies a novel vulnerability in the academic research ecosystem, particularly in peer review and grant evaluation processes.

Ignat explained that the threat involves overwhelming reviewers and editors with a high volume of low-quality submissions generated by AI. Attackers could produce thousands of plausible but fraudulent research papers, grant applications, or conference abstracts. The goal would be to degrade the quality of research, block legitimate submissions, and erode trust in scholarly outputs. The sabotage could target specific fields or institutions.

The Earma conference, which focuses on research management and administration, provided a platform for this alert. He called for the research community to develop new verification methods and safeguards. The warning underscores how AI, while offering benefits, also enables new forms of sabotage when misused, potentially undermining the integrity of the entire research enterprise.