Revolutionizing Assessment: AI Enters Singapore University Classrooms

In the rapidly evolving landscape of higher education in Singapore, artificial intelligence (AI) is making significant inroads into one of the most traditional aspects of academia: grading student work. Lecturers at four prominent public universities—Nanyang Technological University (NTU), National University of Singapore (NUS), Singapore University of Technology and Design (SUTD), and Singapore Institute of Technology (SIT)—are now using AI tools to assess assignments and exams that directly contribute to students' final scores. This shift, which began gaining traction in 2024 and accelerated through 2025, reflects Singapore's position as a global leader in AI adoption, balancing innovation with rigorous academic standards.

The move addresses longstanding challenges in higher education, such as lecturer workload amid dual research and teaching demands, while promising greater consistency in evaluations. However, it also sparks debates on fairness, nuance, and the human element in learning. As universities navigate this hybrid human-AI approach, students, faculty, and policymakers are redefining what effective assessment means in an AI-driven era.

Identifying the Pioneering Universities and Tools

The four universities leading this initiative are NTU, NUS, SUTD, and SIT. Notably, Singapore Management University (SMU) and Singapore University of Social Sciences (SUSS) have opted not to permit AI grading for work impacting final results, citing concerns over reliability.

• NTU and SUTD: Employ Gradescope, an AI-powered platform that scans handwritten answers, groups similar responses using machine learning, and suggests grades based on rubrics. At NTU, it's used for midterm and final exams in select physics and mathematics modules since August 2024. SUTD implemented it in April 2025 for computer science and design school tests featuring short-answer and explanation questions.
• NUS: Utilizes a validated, unnamed AI tool for post-admission English proficiency tests (argumentative essays) since July 2025. The system evaluates content, organization, and language, running double assessments for consistency, with human audits for borderline cases.
• SIT: Deploys AI-Orate, a homegrown chatbot platform trialed in October 2025 with 50 food technology students. It quizzes learners on concepts like microbial reaction rates in code-based projects, adapting questions and recommending grades via transcripts.

These tools represent a hybrid model where AI handles initial processing, but human lecturers retain final authority through mandatory reviews.

How Gradescope Transforms Traditional Grading

Gradescope, developed by Turnitin, leverages computer vision and AI to streamline grading of handwritten or digital submissions. The process unfolds step-by-step: First, the platform digitizes scanned exam papers. AI then clusters similar answers—such as identical mathematical workings—allowing lecturers to grade one exemplar and apply it across the group, with adjustments for variations.

At NTU, Deputy President and Provost Christian Wolfrum notes, “This approach improves consistency and efficiency in grading while ensuring that all grading decisions remain with the instructor.” SUTD's Associate Provost Ashraf Kassim echoes, “AI acts as a supporting partner to the instructor.” Benefits include reduced grading time and fatigue mitigation, crucial in large classes where human graders might vary scores by up to 10-15% due to subjectivity.

In Singapore's competitive academic environment, where precision is paramount, Gradescope ensures equitable treatment, particularly for structured subjects like STEM.

NTU's Implementation: Physics and Maths Modules Lead the Way

NTU pioneered AI grading among the quartet, integrating Gradescope into selected physics and mathematics modules' high-stakes exams from August 2024. This allows lecturers to handle hundreds of scripts swiftly, focusing on pedagogical insights rather than rote marking.

The tool's AI similarity detection has proven invaluable for problem-solving questions, where workings often mirror each other. Students are notified in advance, fostering transparency. NTU undergraduate Ryanna Lee comments, “It’s only fair that AI use by teachers is watched and controlled,” highlighting the need for oversight. Despite this, adoption has smoothed operations, freeing faculty for research pursuits.

SUTD and SIT: Tailored AI for Design and Applied Tech

SUTD's use since April 2025 targets interdisciplinary tests blending short answers and explanations in computer science and design. Gradescope's grouping aids in assessing creative yet formulaic elements, aligning with SUTD's project-based pedagogy.

At SIT, AI-Orate innovates further. In the food technology trial, students coded simulations, then faced chatbot interrogations. Associate Professor Wong Shin Yee explains, “There is a lot of potential, especially for creating highly customisable assessments in large classes.” Undergraduate Lief Chng appreciates the interactivity: “It feels like a second chance to earn marks.” This adaptive format suits applied sciences, reducing assessment from a week to two days.

NUS's Precision Tool for Language Proficiency

NUS employs a rigorously validated AI for English argumentative essays in post-admission tests since July 2025. Trained to mimic expert graders, it scores holistically and repeats for reliability, with humans intervening on edges. Associate Provost Melvin Yap states, “This hybrid approach significantly enhances grading consistency compared to human graders susceptible to fatigue.”

This targets non-native speakers, a key demographic in Singapore's diverse student body, ensuring standardized entry benchmarks.

Key Benefits Driving Adoption

AI grading yields tangible gains:

• Efficiency: Cuts marking time dramatically, vital for overworked lecturers balancing research.
• Consistency: Minimizes human variability, with studies showing up to 20% score alignment improvements.
• Scalability: Handles large cohorts, enabling personalized feedback.
• Adaptivity: Tools like AI-Orate probe deeper understanding.

NUS student Kwan Kaiee observes, “Lecturers in Singapore are overworked.” Amid Budget 2026's AI literacy push, these tools position universities as innovators.

Challenges and Criticisms: Nuances Under Scrutiny

Despite promise, hurdles persist. NUS student Leslie De Souza questions, “How would AI tell me what to improve on?” AI struggles with creative nuance or cultural context, prompting SMU Vice-Provost Venky Shankararaman to caution: “Human judgment must remain central in high-stakes assessments.”

Potential biases in training data and transparency issues fuel skepticism. Universities counter with human oversight and appeals processes. Broader 2025 incidents, like NTU penalizing three students for AI misuse, underscore reciprocal accountability.

Student Perspectives: Trust Building Through Transparency

Reactions vary. Initial wariness gives way to acceptance when informed. SIT's trial shifted views positively, with adaptivity praised. However, calls for audits persist, mirroring global trends where 74% of students stress over AI per surveys.

In Singapore's meritocratic culture, equity is key. Policies mandating disclosure build trust, preparing students for AI-augmented careers via higher ed jobs.

Safeguards and Policies Ensuring Fairness

All four universities enforce:

1. Pre-notice to students.
2. Mandatory human review.
3. Appeal rights.
4. Departmental validation (e.g., NUS).

This aligns with national AI frameworks, promoting responsible use. SMU's mid-2025 working group signals potential expansion.

Photo by Galen Crout on Unsplash

Future Outlook: AI's Expanding Role in Singapore HE

With Budget 2026 allocating for AI skills at IHLs, expect wider adoption. Forums like ST Education Forum 2026 (“AI in Higher Education: Hype or Hope?”) will shape discourse. Challenges like bias mitigation via diverse datasets loom, but hybrid models promise empowered learning.

For aspiring academics, this evolution opens lecturer jobs emphasizing AI literacy. Explore opportunities at Singapore university jobs or rate your professors. As AI redefines assessment, Singapore universities lead, blending tech with tradition for future-ready graduates.

Interested in careers? Check higher ed career advice, higher ed jobs, and university jobs.