Artificial Intelligence in Dental Research: Innovation versus Integrity?

The integration of artificial intelligence (AI), particularly large language models and agentic pipelines, is reshaping dental research. This article examines the opportunities and risks of AI in knowledge production in oral health research. On one hand, AI accelerates literature synthesis, data analysis, and manuscript drafting; on the other, it introduces new risks, including fabricated findings and citations, overconfident inference, and homogenization of research focus. These shifts challenge the mechanisms that maintain scientific credibility such as peer review, reproducibility, and human judgment, especially in fields like dentistry, where small datasets and limited editorial capacity may amplify vulnerabilities. We here identify 3 structural asymmetries: 1) the growing gap between the rapid pace of AI-assisted production and slow human validation, 2) the divergence between AI-generated linguistic fluency and the true robustness of the generated evidence, and 3) unequal institutional capacity to govern AI and the incentives set by the research system for (mis)using AI for research. We conclude with a call for proportional and enforceable governance frameworks, including mandatory AI disclosure standards, provenance tracking, structured adversarial testing of research tools, and interdisciplinary oversight. Moreover, we suggest institutional reform, including revising incentive structures, fostering AI literacy, and enforcing auditability to safeguard research integrity. In the age of machine-assisted content, credibility becomes a scarce resource. Dental science must not merely adopt AI but actively govern it to ensure that its accelerated outputs remain trustworthy, equitable, and clinically meaningful.