The Spark of Controversy at Nanyang Technological University

In the bustling academic landscape of Singapore, Nanyang Technological University (NTU), one of the nation's premier institutions, found itself at the center of a heated debate in mid-2025. Three students enrolled in the module 'Health, Disease Outbreaks and Politics'—offered by the School of Social Sciences and taught by Assistant Professor Sabrina Luk—were awarded zero marks on a critical assignment worth 45 percent of their overall grade. The reason? Alleged misuse of generative artificial intelligence (GenAI) tools, such as ChatGPT, which led to fabricated citations, non-existent references, and inaccurate statistics in their essays.

The assignment instructions were crystal clear from the outset. Briefing slides explicitly stated: 'The use of ChatGPT and other AI tools are not allowed in the development or generation of the essay proposal or the long essay. You will receive a zero mark for the assignment if you are caught using ChatGPT and other AI for writing assignments.' Reminders were issued multiple times, underscoring the instructor's intent to evaluate students' independent research skills, originality, and critical thinking—core competencies in higher education.

Following submission in April 2025, anomalies surfaced during grading: broken web links, invented academic sources, and dubious data points. An investigation ensued, culminating in formal reviews where students could defend themselves. While two admitted to limited AI use—for background research or citation organization—the third claimed ignorance that their reference tool was AI-powered. Despite providing evidence like Draftback time-lapse videos of their writing process and corrected citations, all three faced the full penalty: zero marks, permanent academic warnings, and misconduct notations on their records.

Student Voices: Disputes, Appeals, and Frustrations

The affected students did not go down without a fight. One Year 3 public policy and global affairs major took to Reddit on June 19, 2025, detailing her ordeal anonymously. She argued that citation errors—such as misspelled author names (e.g., 'Lee' instead of 'Li') or wrong dates—were mere typos common among undergraduates, not deliberate fraud. She had even emailed the professor with updated references, receiving acknowledgment but a firm 'no negotiation' in response. Her appeal process involved paying S$40 for a hearing, where procedural hurdles like rescheduling issues left her feeling unheard.

Another student, a Year 4 learner, described using ChatGPT solely to alphabetize citations and gather preliminary insights, insisting no content was AI-generated. Initially docked 10 marks by the professor, the penalty escalated to zero after intervention by the School's Academic Integrity Officer, without a personal hearing. A third recounted a tense review meeting where the professor allegedly raised her voice, dismissing her explanations despite her transparency about using tools like citationmachine.net.

These disputes highlighted deeper tensions. Students felt penalized for minor aids rather than wholesale cheating, with one noting, 'I deeply regret not double-checking my citations and rushing through the process, but I understand the importance of academic integrity.' Appeals yielded mixed results: one had the 'academic fraud' label removed after an expert panel deemed the tool non-GenAI, while others were rejected, including a case upheld by a panel spotting 14 false citations linked to AI hallmarks.

NTU's Stance: Upholding Academic Integrity Amid AI Proliferation

NTU's response was unwavering. A spokesperson emphasized that 'citing non-existent sources undermines the credibility of academic work and constitutes serious academic misconduct.' The university launched investigations in April, allowing formal presentations before imposing sanctions in early May. In a follow-up statement, NTU reiterated its pragmatic approach to GenAI, aiming to equip students for a tech-driven future while enforcing responsibility.

NTU's Policy on the Use of Generative Artificial Intelligence in Research outlines three pillars: responsible, accountable, and transparent use. Researchers (and by extension, students in scholarly tasks) must disclose GenAI beyond basic editing, cite tools fully, and verify outputs for accuracy. Fabricating data or images is outright prohibited, and non-disclosure risks plagiarism charges. For teaching and assessments, resources like the 'Teaching, Learning & Assessment with GenAI' portal guide faculty on designing AI-resilient evaluations and ethical integration.

In this case, the module's no-AI rule aligned with instructor discretion, a common practice to foster authentic skills. NTU documented all misconduct internally, with appeal outcomes final, signaling zero tolerance for integrity breaches.

Defining Generative AI and Its Role in Education

Generative Artificial Intelligence (GenAI) refers to advanced machine learning models—like OpenAI's ChatGPT, Google's Gemini, or Anthropic's Claude—that produce human-like text, code, images, or data from user prompts. Trained on vast datasets, these tools excel at summarizing, brainstorming, and organizing information but often 'hallucinate'—inventing plausible yet false facts, a pitfall evident in the NTU essays.

In higher education, GenAI promises efficiency: aiding literature reviews, drafting outlines, or refining prose. Step-by-step, a student might prompt: 1) 'Summarize key theories on disease politics'; 2) Copy outputs and verify sources; 3) Integrate with original analysis; 4) Cite properly. However, submitting unedited AI text as one's own violates plagiarism norms, while undisclosed use erodes trust.

Singapore's context amplifies this: As a Smart Nation hub, the Ministry of Education (MOE) encourages AI literacy from primary levels, yet universities grapple with balancing innovation and rigor.

Comparative Policies: How NUS and SMU Handle GenAI

NTU's strict enforcement contrasts with peers but fits a shared framework. National University of Singapore (NUS) permits GenAI in assignments if explicitly allowed, mandating disclosure and content ownership. Its Department of History's 2025-2026 guidelines stress ethical use, treating undeclared outputs as plagiarism.

Singapore Management University (SMU) offers a comprehensive Framework for Generative AI Tools, guiding ethical, secure application in teaching and assessments. Faculty decide permissibility, with unacceptable use breaching the SMU Code of Academic Integrity—penalties mirroring NTU's: grade penalties to expulsion.

Across institutions:

• Disclosure is universal for substantive use.
• Instructors set rules per module.
• Verification of AI outputs is students' duty.
• Low reported cheating cases, but experts warn of under-detection.

For deeper insights, explore NTU's GenAI Research Policy or SMU's framework.

Challenges in AI Detection and Enforcement

Detecting GenAI is no simple task. Tools like Turnitin or GPTZero flag stylistic anomalies—repetitive phrasing, unnatural fluency—but falter on edited or human-AI hybrids. False citations, a GenAI signature, require manual audits, as in NTU's expert panels.

Students countered with Draftback, proving keystroke-by-keystroke authorship. Yet, proving non-use remains elusive, fostering 'AI anxiety' among faculty. CNA reported educators viewing blanket detection as a 'lost cause,' advocating redesigned assessments: oral defenses, process portfolios, or in-class writing.

Stakeholder Impacts: From GPAs to Careers

The fallout rippled beyond grades. Zero marks tanked GPAs, risking honors classifications vital for graduate admissions or jobs. Misconduct records, though internal, could surface in references, haunting future prospects in competitive fields like public policy.

Faculty face dilemmas: Enforcing rules preserves standards but risks alienating transparent students. Rice Media critiqued vague boundaries—e.g., is citation sorting 'development'?—urging nuanced policies. Broader implications include eroded trust, with students potentially hiding AI use, undermining learning.

For those eyeing academia, platforms like Rate My Professor offer glimpses into module expectations, while higher ed career advice stresses integrity.

Expert Recommendations and Solutions

Experts advocate evolution over bans. NTU's portal promotes AI-enhanced assessments: process-based grading, AI-inclusive prompts. Ethical use checklist:

• Check instructor guidelines first.
• Disclose transparently: 'Used ChatGPT v4 for initial outline; verified all sources.'
• Cross-verify facts via primary literature.
• Blend with personal voice.
• Practice AI literacy via workshops.

Solution-oriented, universities invest in training: NTU's Inspire initiative, NUS AI modules. Future: AI tutors for feedback, not content creation.

Photo by Masjid MABA on Unsplash

Looking Ahead: AI's Enduring Place in Singapore Higher Education

By 2026, Singapore's Research, Innovation and Enterprise 2030 (RIE2030) allocates billions to AI, positioning universities as hubs. NTU leads with quantum-AI investments, signaling integration over prohibition.

The saga underscores adaptation: Clearer rules, faculty training, student education. As GenAI evolves, so must academia—fostering innovators who wield tools ethically.

Aspiring professionals can leverage this shift via higher ed jobs in AI ethics or university jobs. For career navigation, visit academic CV tips.