NTU upholds zero mark for student initially accused of AI use, citing 14 false citations in essay

SINGAPORE: Nanyang Technological University (NTU) has upheld the decision to award a zero grade to a student whose essay was found to contain multiple instances of false citations. The case gained attention after the student posted on Reddit on 19 June 2025, claiming she had been penalised for using artificial intelligence (AI) in her final module essay. Following her post, NTU convened an appeal panel composed of academic staff and AI experts to review the matter. The student was one of three undergraduates at the university who faced disciplinary action over suspected AI use in academic submissions. On 18 July 2025, NTU released a statement confirming that the appeal panel had identified 14 instances of false citations or fabricated data within the essay. “These were not mere typos,” the panel concluded, adding that some citations referred to non-existent sources.

NTU: “The outcome of the appeal panel is final”

According to NTU, such conduct constitutes a “serious form of academic misconduct,” particularly because citation integrity forms the foundation of academic writing. The university stated that false citations are frequently linked to the output of generative AI tools, which are known to invent or hallucinate references. "False citations of this nature are often due to factual inaccuracies generated by generative AI,” said NTU in the statement. NTU clarified that the findings of the appeal panel are final. All cases of academic misconduct are recorded internally within the university's system. “The outcome of the appeal panel is final. We urge everyone to respect the integrity of the appeals process and refrain from making personal attacks against any individual involved,” said NTU in its public statement.

Student disputes link to AI, cites procedural inconsistency

In response to media queries, the student, who has remained anonymous, expressed frustration over the appeal outcome. She told CNA that during her meeting with university officials, the discussion focused on her citation errors rather than AI usage. “Their entire talk with me during the meeting was on how I can avoid making essay mistakes in the future, didn’t tackle AI usage at all,” she said. According to the student, the decision to uphold the zero was based on the professor’s academic standards, not any confirmed use of AI. The student acknowledged using a citation organiser tool she had found online. It sorted references alphabetically and was not an AI content generator. She also submitted supporting evidence to the panel, including a recording of her writing process using the Chrome extension Draftback. The tool logs keystrokes to demonstrate that she typed the essay manually rather than copying and pasting content from tools like ChatGPT. Despite this, she said the citation organiser was not mentioned again during the appeal. The student contested the classification of her citation issues as fabrications. She claimed the so-called “non-existent sources” resulted from typographical errors. “It’s only non-existent because of the typos. And frankly writing citations wrongly is quite common amongst undergraduates. I just got unlucky,” she said. Among the errors, she noted that she had written an author’s surname as “Lee” instead of “Li” and had misdated two separate references. She declined to release the full list of 14 errors, citing uncertainty over the confidentiality of the appeal documents. Despite having made efforts to defend herself, the student stated she is now choosing to move on. “Frankly speaking I’m too tired to fight any further and exhausted all my avenues. So I think I’ll just move on,” she said. She added that although she felt unfairly treated, she found some consolation in knowing the university had not confirmed AI usage in her case. “At least I know in my heart I didn’t use AI and they couldn’t prove it,” she said.