​In March 2026, I published an annual review article, “Domestic Politics in 2025: South Korea’s Polarized Democracy After Martial Law,” in Korea Observer 57 (1): 1–21. Six of the references in that article do not exist. I did not comment publicly while the journal and my university conducted their reviews. As those reviews have now concluded, I want to update the public record.

I found the six references during a literature search on Google Scholar in the final stages of revising the manuscript. They appeared with known author names, real journal titles, and abstracts that looked legitimate, and they supported background points about Korean politics that are well established in the field. I reviewed what my search returned, but I failed to independently verify the sources. It is possible that these were AI-generated citations that surfaced in my search results, but I do not know for certain. To be clear, I did not use AI to write any part of the article. The argument and analyses are entirely my own. I alone am responsible for verifying every source I use in my scholarship, and I fell short of that responsibility.

Korea Observer’s Editorial Board, in determining the appropriate remedy, consulted guidance from relevant publishing ethics authorities, including Elsevier and the Committee on Publication Ethics, as well as the journal’s own editorial policies. The Board independently replicated my analyses from the materials I provided and confirmed them to be authentic and reproducible. Its AI-detection screening found insufficient evidence of AI-generated authorship. I did not ask to voluntarily retract the article based on the principle that removing the references leaves the argument, analysis, and conclusions intact and supported by other sources. I remained ready to support the decision the journal deemed most appropriate. After its own review, the Board determined that retraction was not warranted. I recognize that this was a difficult decision for the Board as it tried to understand how the references arose. I share that uncertainty, as I cannot fully explain or document how they surfaced either. I have provided the corrected version of the article at my expense, which can be found on the journal's website here. The journal will publish a formal erratum removing the six references as well as any other warranted corrections, together with an Editor's Note on the seriousness of the matter.

Tufts University’s Office of the Vice Provost for Research conducted a separate assessment. I provided all my original manuscript materials for review. It concluded that the available evidence does not support a finding that my actions were intentional, knowing, or reckless and therefore does not warrant the initiation of a research misconduct proceeding. It attributed what happened to honest error, while emphasizing that the matter remains serious, which I fully accept.

I want to apologize to the scholars whose names appeared on work they did not write. I also deeply regret the burden this placed on the journal and on my university, and the difficulty it created for my colleagues, students, and collaborators who have supported me and engaged with my work.

To me, academic integrity is about the standards we strive for in our work, but also how we respond when we fall short. I am making this statement to be transparent and accept full responsibility for my mistake. I have also changed how I work: I now verify all references and search results directly against the publisher of record and retain a copy of every source, including news articles and websites. I plan to share my experience in the classroom to better guide my students to wisely and responsibly navigate a rapidly shifting research terrain.
​
I consider it an enormous privilege every day to be part of a scholarly community devoted to advancing knowledge and teaching the next generation. Part of that is continuing to learn and improve ourselves as scholars. I look forward to focusing again on that important work.