A senior academic at the University of Hong Kong (HKU) has stepped down from a leadership role after an investigation finds that a journal article he co-authored includes fabricated references generated by artificial intelligence, raising fresh concerns about research integrity in the age of AI.
The professor, a prominent scholar in population studies and director of the Hong Kong Jockey Club Centre for Suicide Research and Prevention, resigns as associate dean of the Faculty of Social Sciences following the university’s inquiry. He also withdraws from faculty research committees but retains his directorship.
The case centres on a paper titled “Forty years of fertility transition in Hong Kong”, published in October in the journal China Population and Development Studies. According to the university, the investigation confirms that a number of references listed in the article do not exist and were generated using AI tools.
The use of such tools is not disclosed by the primary author, a PhD student, while the professor is listed as the corresponding author.
Paper withdrawn after publisher probe
The journal’s publisher, Springer Nature, launches its own investigation in mid-November after concerns about the accuracy of the references are raised publicly. It subsequently retracts the paper, stating that it is unable to verify at least 24 cited sources.
In a statement, the publisher says the authors acknowledge using AI tools without disclosure, and the editor-in-chief loses confidence in the reliability of the article. The authors agree to the retraction.
The controversy first surfaces on social media, where users question the validity of multiple citations. Subsequent media checks find that a significant number of listed references—ranging from academic papers to government documents—cannot be traced through their links or identifiers.
Accountability and institutional response
The professor accepts responsibility as corresponding author and requests the paper’s retraction, apologising for the oversight. He says the student used AI to organise references but failed to verify them.
The doctoral student is now subject to disciplinary procedures, the university confirms.
In its statement, HKU reiterates that all researchers are required to meet “internationally recognised benchmarks for quality and ethical conduct” and describes the incident as a breach of academic standards.
AI and academic integrity under scrutiny
The episode highlights growing challenges universities face as generative AI tools become embedded in research workflows. While such tools can assist with tasks such as formatting and literature searches, their misuse—or uncritical reliance—can introduce significant risks, including fabricated or unverifiable citations.
In response, HKU says it will introduce mandatory training and assessment for all researchers on the appropriate use of AI in academic work, signalling a shift towards stricter governance of emerging technologies.
The incident comes at a sensitive moment for the university, which has recently regained its position as Asia’s top institution in regional rankings. Maintaining research credibility, observers suggest, will be critical as institutions navigate both reputational pressures and technological change.
A wider warning for the sector
The HKU case is unlikely to be isolated. As AI tools become more sophisticated and widely used, universities globally are grappling with how to integrate them responsibly into research and teaching.
For institutions, the lesson is increasingly clear: technological adoption must be matched by robust oversight. For researchers, the episode serves as a reminder that accountability ultimately remains human—even when the errors are machine-generated.
- Log in to post comments