MOE has moved AI from a fringe topic to a structural part of Singapore's education landscape.
Key points
AI literacy is embedded in the national curriculum, the Student Learning Space (SLS) platform has AI-powered features, and schools are running pilots with varying levels of ambition.
For students, the practical implication is that AI in school is not a future concern - it is already shaping how you learn, how teachers mark, and what skills the system considers important.
Planning a revision session? Use our study places near me map to find libraries, community study rooms, and late-night spots.
Read in layers
1 second
Read the summary above.
10 seconds
Scan the first few sections below.
100 seconds
Jump into the section that matches your decision.
Quick AI-in-School Map
1 Singapore's Institutional Approach to AI in Education
2 AI Literacy in the National Curriculum
3 The Student Learning Space (SLS) and AI Features
Q: What does How Singapore Schools Are Using AI in 2026 cover? A: An overview of AI adoption in Singapore education: MOE's AI literacy framework, Student Learning Space AI features, school-level pilots, and what JC and secondary students should know about AI in their own education.
TL;DR MOE has moved AI from a fringe topic to a structural part of Singapore's education landscape. AI literacy is embedded in the national curriculum, the Student Learning Space (SLS) platform has AI-powered features, and schools are running pilots with varying levels of ambition. For students, the practical implication is that AI in school is not a future concern - it is already shaping how you learn, how teachers mark, and what skills the system considers important.
Status: Reviewed March 2026 against MOE's published EdTech Master Plan, the SLS AI features documentation, and MOE's guidance on generative AI. School-level pilot details vary and change; check your school's communications for specifics.
Quick AI-in-School Map
If you have...
Take this away
1 second
AI is already part of Singapore schooling.
10 seconds
It shows up through curriculum changes, SLS tools, teacher analytics, and school-level pilots.
100 seconds
Use this guide to separate official MOE features from external AI tools that each school may regulate differently.
Concrete example
A teacher may use AI to create practice questions, while a student using ChatGPT for homework still needs to follow school policy.
Best next step
Check your school's AI guidance before using external tools for assessed work.
1 Singapore's Institutional Approach to AI in Education
Singapore's approach to AI in education is top-down and deliberate. MOE does not leave AI adoption to individual schools or teachers - it sets a national framework, trains teachers centrally, and embeds AI tools into the national learning platform. This is consistent with how Singapore has approached previous technology shifts in education, from the IT Masterplans of the 1990s through the PDLP laptop rollout.
The current framework has four visible layers:
National curriculum changes - AI and data literacy topics embedded in computing and mathematics syllabi
MOE guidance documents - published guidance on responsible AI use, targeting teachers and students
SLS platform features - AI-powered tools within the national learning management system
School-level pilots - individual schools experimenting with AI tutors, marking tools, and AI-assisted lesson planning
Understanding these layers separately is useful because they affect students differently.
2 AI Literacy in the National Curriculum
2.1 Computing and O-Level / A-Level syllabi
The most concrete curriculum change is in Computing. The O-Level and A-Level Computing syllabi now include content on machine learning concepts, data ethics, and AI systems. At the O-Level, students learn how supervised learning works, what training data is, and what bias in datasets means. At the A-Level, the scope includes neural network architectures at a conceptual level.
This is not just technical training. MOE frames AI literacy as part of digital citizenship - understanding how AI systems produce outputs, what their limitations are, and how to evaluate AI-generated information critically. These skills are positioned as foundational for all students, not just those pursuing computing.
2.2 AI in mathematics education
MOE's approach to AI in mathematics is more cautious. There is no AI tool endorsed for use in H2 Maths problem-solving; the emphasis remains on students developing mathematical reasoning independently. However, teachers are using AI to design differentiated practice sets and to analyse where students are making common errors.
2.3 The Digital Literacy Framework
MOE's broader Digital Literacy Framework, which underpins the PDLP and related initiatives, includes AI literacy as one of four pillars alongside cybersecurity awareness, creative digital production, and responsible online participation. Secondary schools are expected to teach AI-related modules within ICT or form teacher time, though the delivery varies significantly across schools.
3 The Student Learning Space (SLS) and AI Features
3.1 What SLS is
SLS (Student Learning Space) is the national digital learning platform managed by MOE. All Singapore mainstream secondary school and JC students have SLS accounts. Teachers assign homework, conduct formative assessments, share resources, and track student progress through SLS. It is not a commercial product - it is a government-managed platform built specifically for the Singapore curriculum.
3.2 AI-powered features in SLS
MOE has been adding AI features to SLS incrementally. The features that have been rolled out or piloted include:
Adaptive learning components. Certain SLS modules - particularly in Mathematics and Science - adjust the difficulty of practice questions based on a student's response pattern. If a student consistently answers a particular type of question incorrectly, the system surfaces more practice of that type. This is a basic form of adaptive learning, not a sophisticated AI tutor, but it represents a shift from static worksheets.
AI-generated feedback on short responses. MOE has piloted AI feedback tools that provide automated comments on short structured responses in SLS - for example, explaining whether a student's answer addresses the question and suggesting improvement areas. Teachers receive these as one input among several; they are not the final assessment.
Analytics for teachers. SLS provides teachers with class-level analytics: which students have not completed tasks, where completion rates are low across a class, which questions most students struggled with. This is data-driven teaching support, and AI-powered insights are increasingly part of how these dashboards work.
3.3 What SLS AI features do not yet do
SLS does not yet have an always-on AI tutor that students can query freely. The AI features within SLS are embedded in specific learning activities, not a general conversational interface. Students who want an AI explanation of a concept still need to go to external tools (ChatGPT, Claude, Gemini) - which puts them outside the school-approved platform and into territory that each school's AI use policy governs.
This is an important distinction. MOE's AI guidance covers the use of both SLS features and external AI tools. The guidance on responsible AI use at MOE's SLS responsible AI page is the authoritative reference for students navigating this.
4 School-Level AI Pilots
4.1 What schools are doing
Individual schools have latitude to experiment with AI beyond the SLS baseline. Observed patterns across Singapore secondary schools and JCs include:
AI-assisted marking pilots. Some schools are trialling AI tools that provide a first-pass mark or comment on structured essays, which teachers then review and adjust. This reduces teacher workload on routine marking while keeping human judgment in the loop. Students may not always be told whether AI-assisted marking was used.
AI literacy enrichment programmes. Many schools - particularly independent schools and SAP schools - have introduced optional or CCA-linked AI literacy modules. These range from workshops on how large language models work to project-based learning where students build simple classifiers.
AI tools for teacher lesson planning. Teachers in Singapore are using AI to design lesson activities, generate differentiated worksheets, and draft rubrics. This does not affect students directly, but it means that some of the materials you receive may have been designed with AI assistance.
Restricted-access AI tutoring trials. A smaller number of schools, in partnership with EdTech vendors, have piloted AI tutoring tools for specific subjects - typically Maths or English. These are school-controlled environments with monitoring, distinct from students using public AI tools independently.
4.2 What is not yet happening at scale
Despite significant attention on AI in education, several often-assumed capabilities are not yet standard in Singapore schools:
There is no school that has fully replaced teacher feedback with AI feedback on major essays or examinations.
AI is not used in SEAB examinations or formal assessments.
There is no national AI tutoring system equivalent to platforms like Khan Academy's Khanmigo operating in Singapore schools.
AI-generated personalised study plans are not a standard offering from most schools.
5 The Smart Nation Context
Singapore's school-level AI activity does not happen in isolation. MOE's education technology strategy connects directly to the Smart Nation initiative, which positions AI fluency as a national economic priority.
The Ministry of Digital Development and Information (MDDI) and the Economic Development Board (EDB) have both emphasised AI skills as central to Singapore's future workforce. MOE's response is to treat AI literacy not as a vocational add-on but as part of general education. The implication is that AI in schools will expand, not contract.
6 What Students Should Know About AI in Their Education
6.1 You are already being assessed in an AI-aware environment
Some schools are using AI to support marking and feedback. Teachers are being trained on AI tools. SLS is collecting performance data that is analysed with AI-powered tools. This does not change what you need to do - demonstrate genuine understanding in your work - but it is worth being clear-eyed about the context.
6.2 AI literacy is becoming a basic expectation
MOE's framing treats AI literacy alongside reading and numeracy as a fundamental skill. Students who graduate from Singapore secondary school or JC without a working understanding of what AI systems are, how they produce outputs, and where they fail will be at a disadvantage in higher education and in the workforce. This is not a distant concern.
6.3 School-level policies on AI use vary significantly
Some schools have detailed AI use policies for different types of work. Others have general guidance that leaves significant discretion to individual teachers. The common thread across all schools is that using AI for assessed work without disclosure is academic misconduct, and that in-examination use is prohibited. Beyond those baselines, the rules differ.
If your school has not communicated a clear AI use policy, ask. Teachers and HODs are better equipped to answer this than they were a year ago.
6.4 The skills AI cannot replace are the ones schools still measure
Examinations - PSLE, O-Level, A-Level - are still sat under conditions that exclude AI. The skills they measure: analytical reasoning, subject-specific knowledge, clear written expression, and the ability to construct a coherent argument under time pressure. These remain the core of what JC is preparing you for. AI can support the development of these skills when used carefully; it does not substitute for them.
7 What This Means for JC Students Right Now
7.1 The SLS AI tools gap at JC level
The SLS AI tools discussed in section 3 - the adaptive learning components, AI feedback features, and analytics dashboards - are deployed at the Secondary school level. As of March 2026, these features do not extend to Junior College. JC students do not have access to FA-Math, SAFA, LEA, or ALS through their school accounts.
This is not a minor footnote. Secondary school students have access to MOE-sanctioned adaptive learning tools built into their national platform. JC students do not. If a JC student wants AI-assisted concept explanations, adaptive practice, or automated feedback, they must go to external tools (ChatGPT, Claude, Wolfram Alpha) that are outside the school-managed environment and subject to each school's AI use policy.
The gap is likely to close - MOE's direction is clearly toward expanding AI features in SLS - but it has not closed yet. JC students are, at present, more dependent on external commercial AI tools than their Secondary school counterparts, and have less institutional support for using them safely and effectively.
7.2 What JC students have instead
What JC students do have through their school environment:
SLS access for resource-sharing, homework submission, and basic analytics (without the AI features available to Secondary students)
Individual school-level AI literacy modules where schools have introduced them (typically as part of CCA, enrichment, or ICT time)
Teacher-directed guidance on AI use in assessed work, which varies significantly across schools
8.1 Education Minister Desmond Lee's February 2026 statements
In February 2026, Education Minister Desmond Lee acknowledged publicly that MOE lacks Singapore-specific data on the correlation between AI tool use and learning outcomes. The admission is significant: despite Singapore's active adoption of AI in education, the Ministry does not yet have local evidence on whether students who use AI tools learn more or less effectively than those who do not.
This is an important counterpoint to the confident institutional narrative around AI in education. The technology has been deployed before the evidence base has been established. For students and parents making decisions about AI use in study routines, this means there is no rigorous Singapore-specific research to cite - only general international findings, which may not translate to the local curriculum context.
MOE's stated response is to continue expanding AI literacy in the curriculum while developing frameworks for measuring AI's impact on learning. The implication is that the evidence base will be built over the next several years, not that the current deployment is paused.
8.2 School-by-school policy variation
A common question among students and parents on forums is whether different schools apply different rules about AI use. The short answer is: yes, significantly.
The baseline is consistent across all schools: AI is not permitted in SEAB examinations, and submitting AI-generated work as your own constitutes academic misconduct. Beyond that baseline, schools vary substantially:
Some schools have issued detailed, subject-by-subject AI use policies that specify exactly what is and is not permitted for different categories of work (homework, class tests, internal examinations, project work)
Others have issued general guidance that leaves the decision to individual teachers
A smaller number have not issued formal guidance at all and are operating on existing academic integrity principles
Independent schools and SAP schools have generally moved faster to develop detailed policies. This is partly because they have dedicated resources for educational technology governance, and partly because their parent communities are more likely to ask about it.
If you do not know your school's AI policy, ask your form teacher or a subject HOD. Most schools have a position, even if it has not been formally communicated.
9 The Parent Perspective
9.1 A different frame from student anxiety
The dominant student concern about AI in education is academic integrity - will using AI get me into trouble? The dominant parent concern, as reflected in KiasuParents discussions, is different: can AI tools reduce the cost of tuition?
This is a materially different question. Singapore parents spend significant amounts annually on private tuition across primary, secondary, and JC levels. If AI tools can replicate some of what a tutor provides - concept explanation, worked examples, practice question generation, essay feedback - that has direct financial implications.
9.2 What AI can and cannot do as a tuition substitute
Honest answer: AI tools can partially substitute for some tuition functions, but not the most valuable ones.
What AI can do that tuition also does:
Explain concepts in plain language, at the student's pace and on demand
Generate additional practice questions on a specific topic
Provide worked solutions to standard problem types
Offer feedback on essay structure and argument logic
What tuition provides that AI cannot:
Diagnosis of the specific gaps in a student's understanding (AI responds to what you ask; it does not identify what you do not know you are missing)
SEAB-aligned feedback that maps student work to marking schemes and examiner expectations
Accountability and structured study sessions that students maintain over time
Singapore curriculum-specific expertise - knowledge of which topics are high-yield for SEAB, how examiners phrase questions, what the mark scheme rewards
The pattern that emerges from honest assessment is that AI tools can reduce the need for tuition at the low end of the value spectrum - concept explanation, additional practice - but do not replicate the diagnostic, SEAB-specific, and accountability functions that distinguish effective tuition from generic study support.
For parents making tuition decisions, AI tools are best understood as a complement that reduces the volume of tuition needed for routine tasks, not as a substitute for subject-specialist academic support during JC.
10 The Responsible Use Baseline
MOE has published guidance on responsible AI use that applies to all students on SLS and when using external AI tools for school-related purposes. The key expectations are:
Follow your school's and teacher's instructions on when and how AI may be used
Do not input personal data or sensitive school information into public AI tools
Be prepared to explain your work and acknowledge AI assistance where required
Understand that AI outputs require critical evaluation - they are not inherently accurate or reliable
