Why tuition centres and private schools need a different approach

Most AI-in-education content in Malaysia is written for public school teachers working inside KPM’s DELIMa ecosystem, or for lecturers inside public university systems with formal continuing professional development (CPD) structures. Tuition centres and private schools sit outside both of those frameworks. As private businesses, they are HRD Corp registrable employers in their own right, which means they can claim training costs the same way any corporate employer does, but the content needs to fit how a tuition centre or private school actually operates: smaller staff numbers, exam-focused teaching, parent-facing communication, and centre owners who are often also the primary decision-makers on training spend.

What AI training looks like for a tuition centre or private school

For teaching staff

Sessions for tutors and subject teachers typically cover using generative AI to draft practice questions and worksheets aligned to SPM, IGCSE, or other exam syllabuses, generating differentiated materials for mixed-ability classes, and speeding up marking-support tasks such as drafting model answers and rubrics. Academic integrity and how to check AI-generated content before using it with students are covered as standard, not as an afterthought.

For administrative and front-office staff

Administrative teams benefit from AI training focused on drafting parent communications, scheduling and timetabling support, and creating marketing content for enrolment drives, tasks that are common in a private education business but rarely covered in public-sector-oriented AI training.

For centre owners and principals

Owners and principals often want a shorter session focused on where AI fits into the business side: content creation for marketing, using AI to draft internal SOPs, and understanding what generative AI is and how it differs from the search tools and basic software staff already use, so they can make informed decisions about what to roll out across the centre.

HRD Corp claims for private education businesses

A registered tuition centre or private school that contributes the Human Resources Development levy can claim eligible AI training under the same SBL (Skim Bantuan Latihan) or SBL-Khas schemes available to any other Malaysian employer. This is a detail many centre owners are not aware of, since HRD Corp claimability is usually discussed in the context of corporate offices or factories rather than education businesses. Centres that have not previously claimed training costs should check their HRD Corp registration status and levy contribution history before applying, since claim eligibility depends on active registration at the time of training.

How this differs from public school and university training

Public school AI training generally runs through the DELIMa platform and KPM’s Digital Education Policy guidance, with content standardised across the national curriculum. University-level continuing professional development for lecturers usually sits inside formal CPD or LDP structures tied to the institution’s own policies. Tuition centres and private schools have neither structure by default, which is exactly why a dedicated programme, rather than a repurposed public-school or university session, tends to fit better. The content, pacing, and claim pathway are built around a private education business rather than a public institution.

How Dr Hariz structures this training

Dr Muhamad Hariz Muhamad Adnan is a Doctor in Artificial Intelligence and Senior Lecturer at the Faculty of Computing and Meta-Technology, Universiti Pendidikan Sultan Idris (UPSI), and an HRD Corp certified AI trainer focused on AI-driven digital transformation in education and the workplace. For tuition centres and private schools, sessions are typically split by role (teaching staff, administrative staff, and owners/principals), run as a half-day or full-day on-site workshop, and delivered in Bahasa Melayu, English, or bilingually depending on the centre’s staff composition.

Frequently asked questions

Can a tuition centre claim AI training under HRD Corp?

Yes, provided the tuition centre is registered with HRD Corp and actively contributing the Human Resources Development levy. The claim process follows the standard SBL or SBL-Khas pathway used by any other Malaysian employer.

Is this the same training given to public school teachers?

No. Public school AI training is generally delivered through the DELIMa platform and KPM guidance. Training for tuition centres and private schools is designed separately, around exam-focused teaching, parent communication, and the business side of running a private education centre.

What topics are covered for administrative staff?

Administrative sessions typically cover drafting parent communications, scheduling support, and creating marketing content for enrolment, rather than classroom-specific AI use.

How long does a typical session run?

Most tuition centres and private schools opt for a half-day or full-day on-site workshop, sometimes split into separate sessions for teaching staff and administrative or ownership staff.

Can the session be delivered in Bahasa Melayu?

Yes. Sessions can be delivered in Bahasa Melayu, English, or bilingually depending on the centre’s preference and staff composition.

Related: AI Workshops for Malaysian Schools (2026) and AI Professional Development for Educators in Malaysia (2026).

Want to bring AI training to your tuition centre or private school? Learn more about AI training for education with Dr Hariz or contact us to discuss your staff and format.

Most guides on AI in education talk about tools in the abstract. This one focuses on what a Kementerian Pendidikan Malaysia (KPM) teacher can actually do this week, inside the systems the ministry already provides, without stepping outside school policy.

Start with DELIMa, not a random AI app

DELIMa is the Ministry of Education’s official digital learning ecosystem, built on Google Workspace for Education. Because it is the sanctioned platform, using its built-in AI features is the lowest-friction way for a KPM teacher to bring AI into a lesson without raising data governance or approval concerns that come with third-party apps. As Google has rolled Gemini for Education into Workspace for Education accounts in participating regions, Malaysian schools using DELIMa gain access to AI features layered directly into the tools teachers already use for documents, slides, and classroom management.

Lesson-level workflows inside DELIMa

Lesson planning with Gemini in Docs

Draft a lesson outline in Google Docs within DELIMa, then use the built-in Gemini assistant to expand a topic into learning objectives, activity suggestions, and formative assessment questions aligned to the syllabus. Review and edit every suggestion; treat the output as a first draft, not a final plan.

Differentiated worksheets in Slides and Docs

Use Gemini to generate two or three versions of the same worksheet at different reading or difficulty levels, which is one of the most time-consuming manual tasks in differentiated instruction. Keep the core learning objective identical across versions so assessment stays comparable.

Feedback drafting, not feedback replacement

For written assignments, AI can draft a first-pass comment bank (common errors, suggested phrasing for feedback) that the teacher then personalises. This keeps the teacher’s professional judgement in the loop while cutting the time spent writing repetitive comments.

Bahasa Melayu prompt examples

DELIMa’s AI tools work in Bahasa Melayu as well as English. Useful starting prompts for BM-medium classrooms include: “Bina rancangan pengajaran harian untuk topik [tajuk] bagi murid Tahun [X], selaras dengan DSKP,” “Hasilkan lima soalan pemahaman aras rendah dan lima soalan aras tinggi untuk petikan ini,” and “Cadangkan aktiviti kumpulan untuk mengukuhkan kefahaman murid tentang [konsep].” Always review AI-generated Bahasa Melayu output for accuracy before use, since translation and terminology errors can slip through.

Working within KPM and Digital Education Policy guidance

Malaysia’s Digital Education Policy sets the direction for how technology, including AI, is meant to support teaching and learning in government schools, with an emphasis on equitable access and teacher readiness rather than unmanaged tool adoption. Practically, this means a teacher’s safest path is to stay inside DELIMa’s sanctioned AI features rather than adopting external consumer AI apps that have not been vetted for student data handling. Where a school wants to go further, such as piloting additional AI tools, this should route through the school’s ICT coordinator and align with existing KPM circulars on digital tool use, not be an individual teacher’s ad hoc decision.

Protecting assessment integrity

AI-assisted lesson preparation is different from AI-assisted student work, and the two need separate rules in the classroom. Three practical habits help: set clear expectations with students about when AI tools are and are not permitted for a given assignment, prefer in-class or oral components for high-stakes assessment where authorship matters most, and design tasks (personal reflection, in-class application, oral defence of written work) that are harder to fully outsource to AI regardless of what detection tools claim to find.

Where this fits with your professional development

DELIMa fluency is a starting point, not the ceiling. Teachers who want a structured pathway from basic AI literacy through to confident classroom integration, including how it maps to LDP and CPD requirements, can see our related guide on AI professional development for educators in Malaysia. For guidance on planning individual lessons with AI in more depth, see AI for lesson planning: a step-by-step guide for Malaysian teachers.

Dr Hariz’s training for schools and IPTA/IPTS institutions is built around exactly this kind of policy-aware, tool-specific approach: practical enough to use on Monday morning, structured enough to satisfy a school’s or ministry’s governance expectations. Programmes are delivered in Bahasa Melayu, English, or bilingually and are HRD Corp claimable where applicable to the institution.

Frequently asked questions

Is DELIMa the only approved AI platform for Malaysian teachers?

DELIMa is the Ministry of Education’s official digital learning platform and the most straightforward, policy-aligned starting point, since it is already sanctioned for use in government schools. Individual schools may pilot additional tools, but this should go through proper school-level approval rather than informal individual use.

Do I need special training to use Gemini inside DELIMa?

Basic use requires little training since the AI features sit inside familiar Google Docs and Slides interfaces, but getting genuinely useful, curriculum-aligned output (rather than generic answers) benefits from structured training on prompt writing and lesson-specific use cases.

Can AI tools in DELIMa access or store student personal data insecurely?

Because DELIMa runs on Google Workspace for Education under KPM’s governance arrangements, data handling follows the ministry’s approved framework, which is the main reason it is preferred over unvetted third-party AI apps for classroom use. Teachers should still avoid entering sensitive personal student information into any AI prompt unnecessarily.

How do I stop students from using AI to cheat on assignments?

Redesign high-stakes assessment to include in-class components, oral explanation, or personal reflection that are harder to fully outsource to AI, rather than relying solely on AI-detection tools, which are inconsistent. Clear, assignment-specific rules on permitted AI use also reduce ambiguity for students.

What is the difference between using AI for lesson planning and using AI in the classroom with students?

Lesson planning use is entirely teacher-side (drafting materials, generating question banks) and carries low risk. In-classroom use with students, such as letting students interact directly with an AI tool, requires additional thought about age-appropriateness, data privacy, and school policy before introducing it.

Want practical, DELIMa-aligned AI training for your school or institution? Contact Dr Hariz to discuss a session for your teaching staff.

This guide explains each pathway, what a good AI professional development programme for teachers and lecturers should contain, how it can be funded, and how to choose a trainer. It is written by Dr Muhamad Hariz Bin Muhamad Adnan, a Doctor in Artificial Intelligence and Senior Lecturer at the Faculty of Computing and Meta-Technology, Universiti Pendidikan Sultan Idris (UPSI), Malaysia’s education university, and an HRD Corp certified AI trainer whose work focuses on AI-driven digital transformation in education. His education programmes are described on the AI for education in Malaysia page.

Why AI professional development is now a priority for Malaysian educators

Generative AI tools (systems such as ChatGPT, Gemini and Microsoft Copilot that produce new text, images and analysis from a prompt) are already in students’ hands. Educators therefore face two demands at once: they need to use AI to reduce their own workload in planning, marking and administration, and they need to teach and assess in a way that stays valid when students have AI access. KPM has issued professional circular guidance on the use of AI in schools, and universities are updating academic integrity and assessment policies. Professional development is how individual teachers and lecturers turn that policy into classroom practice.

Digital transformation in education means using digital tools to change how teaching, assessment and administration actually work, not simply adding a device to an existing lesson. AI professional development is the human side of that transformation: without it, tools are bought but not used.

Pathway 1: In-service training (LDP) for KPM school teachers

Teachers in Malaysian public schools complete annual in-service training, commonly referred to as LDP (Latihan Dalam Perkhidmatan), organised at school, district (PPD) and state (JPN) level. AI topics can be scheduled into LDP in three ways: as a whole-staff workshop delivered by an external trainer, as a school-based professional learning community (PLC) session led by a trained teacher, or as an online module counted towards LDP hours.

For school leaders planning this, the guide to running in-house AI workshops for Malaysian schools sets out a practical whole-staff format. The key design choice is to build the workshop around subjects and tasks the teachers already do (lesson plans, RPH, question banks, parent communication) rather than generic AI demonstrations.

Pathway 2: Continuing professional development for lecturers

Universities and colleges under the Ministry of Higher Education run their own teaching and learning development units, and lecturers are expected to accumulate professional development activity each year. AI professional development for lecturers usually concentrates on three areas: designing assessments that remain valid with AI available, using AI to accelerate research writing and literature review without breaching integrity rules, and building AI literacy into the curriculum of their own courses.

Dr Hariz’s guide to AI for assessment and feedback for lecturers in Malaysia is a useful companion for this pathway.

Pathway 3: HRD Corp claimable AI training for private education employers

Private schools, international schools, private colleges, tuition centres and training providers that are registered with HRD Corp (the Human Resource Development Corporation under the Ministry of Human Resources) can use their training levy to fund AI professional development for teaching staff. To be claimable, the programme must be delivered by an HRD Corp registered training provider and an accredited trainer, and the grant application must be approved before the training date. This makes HRD Corp the most practical funding route for private education institutions, and it applies to both on-site and live online delivery. The employer guide to HRD Corp claimable AI training explains the process step by step.

Pathway 4: Self-directed and platform-based learning

DELIMa (Digital Educational Learning Initiative Malaysia) is KPM’s digital learning platform for schools and gives teachers access to tools that increasingly include AI features. Vendor academies from Google, Microsoft and others provide free AI-for-education modules, and MOOC platforms offer certificates. These are valuable for motivated individuals, but on their own they rarely change whole-school practice. They work best as follow-up after a facilitated workshop, when teachers already have a reason to keep practising.

What a strong AI professional development programme should contain

Whether the audience is primary school teachers or university lecturers, the programmes that produce lasting change share the same structure.

• AI literacy foundation. What Generative AI is, where it is reliable, where it fabricates, and how data privacy applies to student information.
• Hands-on task practice. Participants build their own lesson plans, differentiated materials, rubrics and feedback comments with AI during the session, not afterwards.
• Prompting for educators. Structured prompting techniques adapted to Malaysian curriculum documents and to Bahasa Melayu as well as English.
• Assessment and integrity. Practical redesign of assignments and a clear classroom AI policy that students understand.
• Follow-up and measurement. A 30-day practice plan, a shared prompt library for the school or faculty, and a simple measure of adoption such as hours saved per week.

The practical AI policy framework for Malaysian classrooms can be used as the take-home artefact for the assessment and integrity module.

How to choose an AI trainer for educators

The trainer market for AI in Malaysia is dominated by corporate and technical providers. For educators, look for three additional qualities: genuine teaching experience so the trainer understands classroom constraints, familiarity with KPM and higher education policy context, and a formal AI qualification so the content is accurate rather than a repackaged tool demo. Dr Hariz combines all three, as a university lecturer at UPSI, a Doctor in Artificial Intelligence and an HRD Corp certified trainer, and he delivers programmes for schools, universities and private education groups across Malaysia.

Frequently asked questions

Can AI training count towards a teacher’s LDP hours in Malaysia?

In-service training organised or endorsed by the school, PPD or JPN is generally recorded as LDP. Confirm the recording arrangement with your school’s LDP coordinator before the session so that attendance and hours are captured correctly.

Is AI professional development for teachers HRD Corp claimable?

For employers registered with HRD Corp, such as private schools and colleges, yes, provided the programme is delivered by an HRD Corp registered provider and an accredited trainer, and the grant is approved in advance. Public school teachers under KPM are funded through KPM’s own training mechanisms rather than HRD Corp.

How long should an AI workshop for educators be?

A one-day (six to seven hour) hands-on workshop is the most common format and is enough for teachers to leave with usable lesson materials. Two-day programmes allow deeper work on assessment redesign and AI policy. Half-day sessions work as awareness briefings but rarely change practice on their own.

What should educators learn first: ChatGPT, Gemini or Copilot?

The tool matters less than the workflow. A good programme teaches transferable prompting and verification skills using whichever tool the institution has already approved, and shows how the same approach applies across tools.

Can the training be delivered in Bahasa Melayu?

Yes. Dr Hariz delivers AI professional development in Bahasa Melayu, English or bilingually, depending on the audience.

Next step

If you are planning AI professional development for your school, faculty or education group, contact Dr Hariz with your audience size, preferred dates and whether you need an HRD Corp claimable arrangement. You will receive a proposed agenda and quotation tailored to your institution.

Related: AI Training for Tuition Centres and Private Schools in Malaysia (2026)

The distinction matters because most universities in Malaysia buy AI training the way they buy corporate training, then wonder why it does not stick. A commercial two-day generative AI course written for a bank does not answer the questions an academic actually has: what happens to assessment integrity, how do I supervise a student who used AI in their literature review, and what am I allowed to put into a chatbot when the data belongs to the university.

Who this is for inside a Malaysian university

Institutional AI training is usually commissioned by one of four offices, and each one wants something different:

• The Centre for Academic Development or equivalent teaching and learning centre, which owns continuing professional development for lecturers and typically wants classroom-facing content.
• The Deputy Vice-Chancellor (Academic) office, which owns policy, assessment integrity and programme-level learning outcomes.
• The research management centre, which cares about AI in the research workflow, publication ethics and postgraduate supervision.
• The registrar or human resources division, which is often the budget holder and thinks in terms of staff competency frameworks and annual training hours.

Getting these four aligned before the first session is the single biggest predictor of whether the programme changes anything. When they are not aligned, the institution ends up with a well-attended awareness talk and no change in practice the following semester.

What a staged rollout actually looks like

Stage 1: a shared AI literacy baseline

AI literacy is the ability to understand what an AI system does, judge when its output can be trusted, and use it responsibly within a professional context. At institution level this means every academic staff member should be able to explain, in plain terms, what generative AI is (a system that produces new text, images or code by predicting patterns learned from training data), why it produces confident errors, and where the institution draws its lines.

This stage is deliberately broad and short. It is the only stage that everyone attends together, and its purpose is to stop the two failure modes at either extreme: staff who ban AI outright without understanding it, and staff who treat model output as authoritative.

Stage 2: role-specific tracks

After the baseline, the programme splits. A teaching track covers lesson design, materials preparation and the redesign of assessment so that it remains meaningful when students have AI access. A supervision and research track covers literature screening, methodology support, the limits of AI-assisted writing under journal and university policy, and how to have an honest conversation with a postgraduate student about disclosure. An administration track covers document drafting, minutes, data handling and the practical governance questions around institutional data.

Stage 3: policy and embedding

Training without policy produces confident individual practice and institutional inconsistency. The third stage is where the institution writes down what is permitted, what must be disclosed, and what is prohibited, then embeds that into course outlines, assessment rubrics and postgraduate handbooks. This is also where a train-the-trainer cohort matters: a handful of internal champions who can run the baseline session for new staff next year, so the institution is not buying the same workshop again indefinitely.

Why the trainer background matters here

Universities are one of the few buyers where the trainer being an academic is not a nice-to-have. A trainer who has never sat on an examination board, never supervised a thesis and never had to defend an assessment decision cannot credibly advise on assessment integrity or supervision disclosure.

Dr Muhamad Hariz Bin Muhamad Adnan holds a Doctor in Artificial Intelligence and is a Senior Lecturer at the Faculty of Computing and Meta-Technology, Universiti Pendidikan Sultan Idris (UPSI). He is also an HRD Corp-certified AI trainer in Malaysia, with a working focus on AI-driven digital transformation in education and the workplace. That combination means institutional sessions are delivered by someone who teaches, supervises and assesses inside a Malaysian public university, not only someone who trains about it.

Budget and funding routes

Malaysian institutions typically fund AI training through one of three routes. Public universities often draw on their own staff development allocation or a centre-level CPD budget. Private institutions and university-linked companies with employees registered under the Human Resources Development Corporation (HRD Corp) may be able to use the levy, since HRD Corp claimability depends on employer registration and programme approval rather than on the sector. Grant-funded and project-linked training is a third route where AI upskilling forms part of a wider digital transformation initiative.

Whichever route applies, scoping the cohort honestly saves money. An institution with 400 academic staff does not need 400 people in the same advanced generative AI workshop. It needs 400 people through a short shared baseline and perhaps 80 through the deeper role-specific tracks.

Frequently asked questions

How long does an institutional AI training programme take?

A realistic staged rollout runs across one to two semesters. The baseline stage can be delivered in a half day per cohort, role-specific tracks usually take one to two days each, and the policy and embedding work runs alongside rather than after.

Can we start with just one faculty instead of the whole university?

Yes, and it is often the better approach. A single faculty pilot lets the institution test the content, gather feedback from staff who will be honest, and refine the policy language before committing the full staff development budget.

Is AI training for lecturers different from AI training for corporate staff?

Substantially. Corporate programmes optimise for productivity and workflow speed. Academic programmes must additionally address assessment validity, academic integrity, research ethics and supervision, none of which appear in a standard corporate curriculum.

Does this cover students as well?

Staff training and student training are separate programmes with different objectives, but they should share a vocabulary and a policy. Training staff first is the usual sequence, because staff are the ones who will set and enforce the rules students follow.

What should we ask a provider before appointing them?

Ask whether the trainer has supervised postgraduate students, whether the content will be adapted to your existing academic integrity policy, whether a train-the-trainer component is included, and what the institution is left with after the last session ends.

Next step

If your faculty, centre or university is scoping AI upskilling for academic staff in 2026, the fastest way to get a realistic plan is a short scoping conversation about cohort size, existing policy and which of the four offices above is driving the request. Get in touch here to discuss an institutional programme.

Related reading: AI for education in Malaysia, corporate AI training in Malaysia, and building AI literacy in Malaysian schools and universities.

What ChatGPT can (and can’t) do for teachers

ChatGPT is a writing and reasoning assistant. It is genuinely good at drafting, rephrasing, summarising, brainstorming, and explaining concepts at different levels. It is not a source of truth: it can state wrong facts confidently, invent citations, and reflect bias. The right mental model is a fast, tireless teaching assistant whose work you always review before it reaches students.

That single principle — draft with AI, decide as a teacher — is what separates effective classroom use from risky shortcuts.

Seven practical ways teachers use ChatGPT

The highest-value uses are the repetitive tasks that eat into a teacher’s evenings. In my workshops, these seven come up most often: drafting lesson plans aligned to learning outcomes; generating differentiated worksheets for mixed-ability classes; creating practice questions and quizzes with answer keys; writing rubric-based feedback comments to refine by hand; explaining a hard concept three different ways for struggling students; producing bilingual Bahasa Malaysia and English versions of materials; and brainstorming engaging hooks, analogies, or project ideas for a topic.

Each of these still ends with the teacher editing the output to match their class, their standards, and their voice.

Ready-to-use ChatGPT prompts for the classroom

Good prompts are specific about the subject, level, length, and format you want. A few starting points you can adapt:

“Act as a secondary school science teacher. Create a 40-minute lesson plan on photosynthesis for Form 2 students, with learning outcomes, a starter activity, a main activity, and an exit question.”

“Generate 10 multiple-choice questions on the water cycle for 10-year-olds, with an answer key and one common misconception explained per question.”

“Rewrite this paragraph at three reading levels — beginner, intermediate, and advanced — keeping the key facts the same: [paste text].”

“Give me three ways to explain long division to a student who is struggling, using everyday examples.”

The pattern is always the same: give role, audience, topic, and the exact output you need.

Keeping AI use safe, fair, and honest

Responsible use is non-negotiable in a classroom. Three habits keep it safe. First, protect privacy: never paste students’ full names, identification numbers, or sensitive personal data into a chatbot. Second, check every fact and figure before it reaches students, because the model can be wrong. Third, be transparent: set clear rules with students about when and how they may use AI, and model honest use yourself by acknowledging AI assistance where relevant.

These are exactly the guidelines I help schools turn into a simple, written classroom AI policy.

Frequently asked questions

Is ChatGPT free for teachers?

There is a free tier that covers most everyday teaching tasks. Paid plans add newer models and higher limits, but most teachers can start effectively on the free version.

Is it cheating for teachers to use ChatGPT to plan lessons?

No. Using AI to draft lesson materials is a productivity tool, like using a textbook or a worksheet generator. The teacher remains responsible for the content’s accuracy and suitability. Cheating concerns relate to students passing off AI work as their own, which is why clear classroom rules matter.

Can ChatGPT grade students’ work?

It can draft feedback and suggest rubric-based comments, but it should not be the final grader. Marks and fairness judgements must stay with the teacher, who understands each student’s context.

How do I start using ChatGPT as a teacher in Malaysia?

Start with one low-risk task, such as drafting a single lesson plan, and review the output carefully. Build a small library of prompts that work for your subject, and set clear rules for student use. A short hands-on workshop shortens the learning curve considerably.

Learn to use AI confidently in your teaching

If you or your institution want hands-on training in using ChatGPT and other AI tools responsibly, I run practical workshops for teachers, lecturers, and education leaders across Malaysia. You can read my broader guide on AI in education in Malaysia and how I use AI in my university courses, explore my AI in education training, or get in touch to arrange a session for your school.

Written by Dr Muhamad Hariz Bin Muhamad Adnan — Doctor in Artificial Intelligence, Senior Lecturer at UPSI, and an HRD Corp–certified AI trainer in Malaysia.

Related: How to Use AI in the Classroom in Malaysia (DELIMa Guide)

Why Lecturers Need To Take AI Seriously Right Now

Artificial Intelligence is no longer a future topic. It is a present reality that is reshaping how students learn, how research is conducted, and how graduates will work. As a Senior Lecturer at the Faculty of Computing and Meta-Technology at Universiti Pendidikan Sultan Idris (UPSI), I have seen firsthand how students who understand AI gain a significant advantage in their academic performance, final year projects, and careers.

But here is the challenge: most university lecturers in Malaysia were never trained on how to teach with AI. They were trained in their subject matter, whether it is education, business, engineering, or science, and AI was never part of their initial preparation. That gap is what I try to bridge through my courses, research, and training programmes.

The Courses I Teach That Involve AI

At UPSI, I teach several courses that directly involve AI and related technologies. These include:

• Artificial Intelligence – a core course covering AI fundamentals, machine learning concepts, and practical applications in real-world contexts.
• Generative AI and ChatGPT – a course I designed specifically to help students understand and responsibly use large language models and generative tools.
• Cloud Computing and Data Science – where AI tools for data analytics and cloud-based deployment are covered alongside technical foundations.
• Technopreneurship – where students explore AI-driven business models, startup thinking, and digital innovation.

Across all of these, I combine theoretical grounding with hands-on, project-based learning so students leave with both understanding and practical skills.

How I Structure AI Into My Teaching

Integrating AI into teaching is not just about using AI tools during lessons. It requires a deliberate structure that balances content knowledge, critical thinking, and practical application. Here is how I approach it:

Step 1: Building Foundational Understanding First

Before students touch any AI tool, I ensure they understand the underlying concepts. What is a model? What is training data? What are the limitations of AI? This prevents shallow adoption where students think AI is magic and instead helps them develop sound judgement about when and how to use it.

Step 2: Introducing Tools With Guided Practice

Once the foundation is set, I introduce AI tools with structured tasks. For example, in my Generative AI course, students do not just use ChatGPT freely. They begin with guided prompt design exercises where they compare outputs, analyse quality, and reflect on what makes a prompt effective. This teaches them to think critically about AI output rather than accept it uncritically.

Step 3: Project-Based Learning With Real Problems

The most impactful part of my AI courses is project-based learning where students apply AI to real problems. Past projects have included AI tools for education accessibility, AI-driven data dashboards for community organisations, and generative AI prototypes for local business challenges. This grounds AI learning in genuine context rather than abstract exercises.

Step 4: Ethics and Responsibility as a Running Thread

Throughout every course, ethical use of AI is not a standalone lecture. It is a running thread. We discuss bias in AI models when studying machine learning. We discuss academic integrity when covering generative AI. We discuss privacy and data governance when working on cloud-based AI projects. This ensures students develop a responsible AI mindset from day one.

Gamification and Immersive Learning in AI Education

One innovation I have introduced into my AI teaching is gamification. Learning AI concepts can feel dry or intimidating, especially for students who do not come from a technical background. By introducing game-like elements such as challenges, points, leaderboards, and team-based problem solving, I have seen significant improvements in engagement, attendance, and performance.

I have also experimented with immersive virtual learning environments where students explore AI concepts in simulated, interactive spaces. This kind of experiential learning has been recognised with teaching innovation awards at UPSI and reflects my commitment to making AI education both rigorous and enjoyable.

How I Handle Generative AI and Academic Integrity

One of the most common questions I get from fellow lecturers is: how do you manage ChatGPT and academic integrity? My honest answer is that banning AI tools is not a sustainable solution. Instead, I focus on designing assessments that require genuine understanding, original thinking, and personal reflection, elements that AI cannot substitute.

I also have open, direct conversations with students about responsible use. I make clear that AI is permitted as a thinking tool, a drafting aid, or a research assistant, but that the final work must represent the student’s own analysis and voice. When students understand the reasoning, they are far more likely to use AI ethically than when they are simply told not to use it.

Professional Development: Helping Other Lecturers Do The Same

Beyond my own courses, I conduct professional development workshops for lecturers at universities, polytechnics, and teacher education institutions across Malaysia. In these programmes, lecturers experience the same hands-on, reflective approach I use with students. They leave with:

• A clear understanding of which AI tools are relevant to their subject area.
• Ready-to-use lesson ideas that incorporate AI without disrupting their existing syllabus.
• A practical framework for managing AI use in assessments and student projects.
• Confidence to continue experimenting with AI in their own classrooms.

These workshops are available for individual faculties, full universities, or cross-institutional events, and can be delivered in Bahasa Malaysia or English.

What Lecturers Usually Ask Me

Do I need a technical background to use AI in my teaching?

No. The AI tools most relevant to teaching, such as ChatGPT, Microsoft Copilot, and AI lesson planning tools, do not require programming knowledge. What matters most is understanding how to frame tasks, evaluate outputs, and guide students in responsible use. I cover all of this in my training programmes for non-technical lecturers.

Will AI make lecturers redundant?

No. AI can automate certain tasks like generating draft materials or providing initial feedback, but it cannot replace the mentoring relationship, contextual judgement, and human connection that define great teaching. Lecturers who learn to use AI well will be more effective and more impactful, not less relevant.

How do I start integrating AI into my courses right now?

Start small. Pick one task in your teaching, such as drafting discussion questions, creating a rubric, or preparing a case study, and experiment with using a generative AI tool to support it. Evaluate the output critically, refine it to match your standards, and reflect on what worked. This iterative approach is how I began, and it remains the most sustainable path to confident AI integration.

Where can I learn more about AI in education from Dr Hariz?

You can explore my research, publications, and training programmes on this website. I also welcome inquiries for speaking engagements, faculty workshops, and curriculum advisory work at universities and educational institutions in Malaysia and internationally.

Start Your AI In Education Journey Today

Whether you are a lecturer just beginning to explore AI or an academic leader designing institution-wide AI literacy programmes, I would be glad to support you. My approach is always practical, contextualised to Malaysian higher education, and grounded in research and real classroom experience.

Reach out through the contact page on this website, or connect with me on LinkedIn to discuss how we can bring responsible, impactful AI into your teaching and institution.