Dr. Daniel Jeffrey Koch’s Post

Digital Education and AI: Securing Malaysia’s Future Generation The written reply by the Minister of Education dated 11 February 2026 regarding the readiness of schools under the Ministry of Education Malaysia (KPM) to face the era of digital transformation and Artificial Intelligence (AI) is both encouraging and reassuring. As the Member of Parliament who raised this question, I welcome the Ministry’s commitment as it enters a critical transition phase in driving the digital transformation of our national education system. This is no longer an option — it is an urgent necessity if Malaysia’s younger generation is to remain relevant and competitive in the global arena. The implementation of the Digital Education Policy (DPD) and the strong emphasis placed within the Malaysia Education Blueprint 2026–2035 clearly demonstrate the Government’s seriousness in integrating AI systemically, strengthening blended learning approaches, and building a future-ready digital ecosystem. This signals a clear direction — we do not want our children to merely consume technology, but to become innovators and leaders in it. The data presented is promising. Ninety-one percent of KPM educational institutions have been equipped with digital devices, involving more than 244,000 units nationwide. This reflects a structured and large-scale effort. Furthermore, the expansion of device leasing to an additional 1,769 schools beginning in the third quarter of 2026 will ensure more equitable access for students across the country. The Hybrid Classroom Pilot Project (Smart Class Initiative), currently implemented in 110 selected schools and set to expand to 400 additional institutions involving 2,000 classrooms by the end of 2026, is a strategic move. While I strongly support these initiatives, I also wish to emphasise that the success of digital transformation does not rest solely on infrastructure and devices. Teacher training, student readiness, parental support, cybersecurity awareness, and ethical AI usage must be prioritised. Artificial Intelligence should serve as a tool to enhance learning — not replace the indispensable role of teachers as mentors and moral guides. Malaysia has tremendous potential to emerge as a progressive leader in digital education. With careful planning, consistent implementation, and continuous monitoring, we can nurture a generation that is digitally literate. As a Member of Parliament, I will continue to support and monitor the implementation of these initiatives to ensure that no school — including those in semi-urban and rural constituencies such as Batu Gajah — is left behind in this transformation. The future of our nation begins in today’s classrooms. And the classroom of tomorrow must be digitally empowered, AI-driven, and value-based. YB Dr. V. Sivakumar Member of Parliament for Batu Gajah

We seem to be going through wave after wave of educational technology (AI being the latest), and each time a new tool arrives promising to fix schooling, transform learning, or finally prepare students for an uncertain future, the cycle begins again: excitement, experimentation, early adoption, and ultimately a kind of drift, where the novelty wears off and the old problems stubbornly remain. I don’t see this pattern as a sign that technology is inherently harmful, nor do I think educators are simply gullible or in love with the shiny new object. What I increasingly think is that we are missing a more fundamental question. One that, if asked early enough, might guide practice differently and make the interventions that follow genuinely meaningful rather than merely fashionable. Too often we start with the technology itself: first we notice something new, we marvel at its potential, and only afterwards we look for a place to put it in education. We ask, “Isn’t this amazing? Surely this could help in our schools somehow,” and we begin looking for ways to make it fit. But that is putting the cart before the horse. If we want to make progress as a system, i.e., if we want to help young people learn in ways that endure, that transfer and that prepare them for the complexity of life, we need to begin by asking what problems we are genuinely trying to solve, not what tools might offer the most appealing promises. When we invert that order, novelty becomes a proxy for progress, and momentum replaces meaning. Part of the cost of this inversion is conceptual confusion. If we are not clear about what we expect from K–12 education, then anything may start to look like a necessity. Engagement metrics are mistaken for learning. Adaptive paths are mistaken for understanding. Speed and ease become the standards by which we judge success. And when that happens, we are no longer solving educational problems so much as managing motion. But what are we actually trying to solve? That question deserves more attention. Educational programmes and initiatives have long embraced ideas such as personal development or preparation for work, but rarely do we define what those actually mean in terms of the cognitive architecture of learners, especially learners who are still novices in their own learning processes. If we think that preparing students for life means equipping them with the skills of today, we risk committing ourselves to a kind of rat race, forever chasing what is current without ever thinking about what is durable. The tools that seem indispensable today may be obsolete tomorrow, and the world students will engage with as adults will almost certainly demand capacities we cannot yet foresee. What will remain constant (as it has for every

Shanghai American School recently hosted its second Seizing the Moment: Inspiring Learners Today to Transform Tomorrow conference, welcoming more than 600 educators from SAS and peer schools across the region for two full days of professional learning and collaboration. With over 100 breakout sessions led by 83 SAS presenters, more than 20 student presenters, and over 5,000 total hours of professional learning, the conference reflected SAS’s deep and ongoing commitment to educator growth. The conference featured three keynote speakers who each offered their perspectives on the evolving relationship between humans, technology, and education. Educational innovator and CEO of Designing Schools Dr. Sabba Quidwai challenged educators to view artificial intelligence not as a replacement for thinking, but as a partner in the learning process. “AI should not replace people. It should elevate them,” Quidwai said. “When we treat AI as a teammate rather than a shortcut, we create space for deeper thinking, creativity, and human connection.” Kevin Pereira, an AI consultant and educator from Blu Artificial Intelligence, urged teachers to look beyond finished products and instead pay attention to the thinking behind them. “In the future, what will matter most is not just what students create, but how they create it,” Pereira explained. “We need to help them understand their process, their decision making, and how to use these tools responsibly.” Joining the conference virtually from the World Economic Forum in Davos, an alumnus from the Class of 2010 and Entrepreneur in Residence at Strikeforce, Max Song drew on his experience as an entrepreneur and investor in technology and sustainability. “The world my classmates and I entered after graduation looks very different from the one students are entering now,” Song said. “We have to continue to change, innovate, and push the boundaries and prepare our kids for the world that they will enter as adults.” Teachers facilitated more than 100 workshops and breakout sessions across both campuses. Topics included classroom applications of AI, design thinking, interdisciplinary learning, assessment, and student wellbeing. More than 20 current students led breakout sessions, presented projects and participated in panel discussions, sharing how technology shapes their learning and how they envision their future in an AI-influenced world. For Renee Couturier, Head of Educational Programs, the conference reflects a broader commitment to professional learning. “SAS is a learning organization,” she said. “These days reflect who we are and what we value. We are fortunate to have the time, expertise, and resources to create experiences like this for our educators.” This conference affirmed that while tools and technologies will continue to evolve, SAS’s commitment to learning remains constant.

🌪️Education as a Low-Entropy Engine: Building Character and Knowledge Together Most attempts to change the world fail for a structural reason: the world operates as a high-entropy system, noisy, fragmented, and resistant to coherent transformation. Energy poured directly into such a system dissipates almost immediately. A more effective approach emerges in certain educational environments: the deliberate creation of low-entropy zones. *What Low-Entropy Education Creates Within these zones, classrooms, seminars, mentoring circles, noise drops and alignment rises. Attention converges. Values stabilize. A small group becomes a coherent node where energy can actually do work. In such a node, education does more than transmit knowledge. It becomes a controlled environment where character can be shaped alongside intellect. Discipline, clarity, resilience, and ethical grounding grow in the same structured space where analytical skills and domain expertise are built. This is the crucial difference. High-entropy educational systems focus solely on information transfer. Low-entropy systems cultivate both competence and character simultaneously. *How It Propagates Once a low-entropy educational zone stabilizes, it begins to radiate outward. Students carry both competence and character into their workplaces. Ideas propagate across networks. Standards shift. The initial investment of energy does not vanish; it multiplies. This multiplication happens because character provides the stability that allows knowledge to be applied with consistency and integrity. Without character formation, even brilliant technical training produces erratic results. *The Strategic Insight Meaningful change does not come through system-wide confrontation. It comes through local pockets of order that gradually reshape the larger field. Consider Warren Buffett's decades of market outperformance not through sophisticated models but through disciplined principles. Or lean AI startups achieving comparable results to billion-dollar infrastructure through smarter design. The pattern repeats: focused, low-entropy approaches outperform scattered, high-entropy ones. Education follows the same physics. The most powerful educational interventions create tight, coherent learning environments where both intellectual capacity and character strength develop together under structured conditions. *The Larger Implication Education, at its best, is a low-entropy engine, quietly rewriting the world by cultivating minds and forming character, one stable node at a time. The question for leaders, educators, and institutions: Are you building high-entropy information distribution systems, or low-entropy character and knowledge formation environments? The difference determines whether your energy dissipates or compounds. #Education #Leadership #CharacterDevelopment #SystemsThinking #Strategy #Learning

As higher education institutions navigate AI-integrated learning environments, the need for structured, defensible assessment practices has never been greater. The Pax Assessment Framework (PAF) has already been applied to hundreds of written assignments across disciplines and has consistently supported fair, transparent, and defensible grading decisions. Applying the framework manually has proven academically rigorous, but time-intensive. The PAF Grading platform was designed to address this challenge by embedding the framework into a structured and efficient online workflow, making it significantly faster to produce consistent, well-documented grades accompanied by clear written feedback. PAF Grading is appropriate for a wide range of assessments across disciplines and academic levels, from introductory undergraduate work to advanced graduate research writing. The platform evaluates both the submitted work and the student’s learning journey (from their AI conversations). It supports rubric-based grading and generates draft feedback for instructor use. Importantly, PAF Grading is a decision-support tool, not an automated grading system. It supports academic judgement - it does not replace it. All recommendations and draft feedback require instructor review, editing, and confirmation - no grade is ever assigned automatically. The platform is built to support defensible, auditable assessment practices and to align with accreditation and quality assurance standards emphasizing fairness, transparency, and documented assessment criteria. We are currently seeking beta testers to use the live platform and provide feedback. Beta users receive free, unlimited access to PAF Grading during their testing period to grade their written assignments, generate and download justification reports and grading tables. We are particularly interested in feedback from educators, academic leaders, and edtech professionals exploring AI-integrated assessment systems. This beta phase will directly inform the broader rollout of PAF Grading. If you would like to be included, please DM me or click the link in the comments to apply.

AI in education is no longer a discussion. It’s the new baseline. With Google Gemini launching its SAT practice experience, one thing is clear: 👉 The SAT prep ecosystem is evolving — fast. But here’s my honest view as a founder who works closely with teachers every single day: AI alone will not improve SAT scores. Great teachers, empowered by the right systems, will. That belief is exactly why we’ve built Zeal Educators the way we have. 👀 First look: Zeal.AI (currently under development) We’re now building Zeal.AI — our in-house academic intelligence system — and sharing an early first look. Zeal.AI is not a chatbot and not a shortcut. It’s being designed as a pedagogy-first academic co-pilot for teachers. The goal is simple: help excellent SAT teachers teach better, faster, and more effectively. Why we’re building Zeal.AI (for teachers, not instead of them) Zeal.AI is being developed to support educators with: High-quality, SAT-aligned content curation Lesson, drill, and assessment design support Consistency, rigor, and clarity across classrooms It’s continuously shaped with inputs from academicians and experts from IIT, Harvard, Stanford, MIT, and other leading institutions — because pedagogy must always come before technology. SAT prep at Zeal: built for real outcomes Alongside this, Zeal already operates an advanced SAT portal that focuses on outcomes, not just practice: Accurate SAT score prediction Clear skill-wise and section-wise diagnostics Actionable data showing where students actually lose points Progress tracking aligned closely with real SAT performance To deepen practice exposure, we’re also partnered with The Princeton Review, giving students access to unlimited, high-quality SAT practice — while teachers focus on strategy, clarity, and execution. The result? A true 360° SAT preparation system where teaching effort translates into measurable score gains. What truly differentiates Zeal Technology matters — but who builds it, and why, matters more. Zeal Educators is led by an Ivy League alumni founding team with deep exposure to global education markets. That perspective helps us build systems where: Teachers feel empowered, not constrained Academic quality is non-negotiable Outcomes — not hype — define success A note to SAT teachers If you’re an SAT educator who: Cares deeply about student outcomes Believes AI should support teachers, not replace them Wants to be part of a high-standard, future-ready academic ecosystem I’d genuinely love to connect. The future of SAT prep will be built by strong teachers + intelligent systems — together. Feel free to comment or DM me. Let’s explore how we can collaborate meaningfully. — Anirudh Das Founder, Zeal Educators #SATTeachers #SATPrep #AIinEducation #EdTechCollaboration #ZealEducators #FutureOfLearning

The future of SAT prep is being reshaped — and it’s being built with teachers at the center. Our Founder, Anirudh Das, shares his perspective on why AI alone isn’t enough, and how strong pedagogy + intelligent systems create real academic outcomes. 👀 First look at what we’re building: Zeal.AI — our in-house academic intelligence system (currently under development), designed as a pedagogy-first co-pilot to support teachers with lesson planning, assessments, and data-driven insights. At Zeal Educators, our approach is clear: AI as an enabler. Teachers as the core. Outcomes as the measure. 👇 Read the full post by our Founder and join the conversation on the future of AI in SAT preparation.

Why EdTech Isn’t “Mostly Useless.” A Global Response to a Misleading, US-Centric Verdict The Economist's article "Edtech is profitable. It is also mostly useless" reads like a sweeping verdict on education technology. But in reality, it's a US-centric critique presented as universal truth. The issue is, when Western media equates local failures with global conclusions, it distorts the discourse that shapes funding and policy worldwide. As both an EdTech creator in LMICs and a parent in the US, I agree with parts of the article, but strongly disagree with its universal verdict. ➡️ The question the article never asks Before declaring "EdTech" a failure, we need to ask:  What job is it doing? What problem is it solving? In what context? The same label describes radically different interventions: -> In Well-Resourced Schools (highly skilled teachers, smaller classes) • EdTech replaces human instruction & peer interaction • Result: flat or negative returns; it crowds out what already works -> In Under-Resourced Systems (less-skilled teachers, huge classes) • EdTech substitutes for inadequate instruction • Result: can deliver meaningful gains when filling a crucial gap ➡️ Where EdTech can work: under-resourced systems In much of the world, the big education problem is that there simply aren’t enough adequately-trained teachers to provide quality education, resulting in a learning crisis. Here, appropriately designed technology can help substitute for missing human capacity and improve learning outcomes. Evidence from LMICs is growing: EIDU in Kenya, Mindspark in India, and our own work at Beaj in Pakistan have demonstrated measurable gains when technology fills real gaps. But this field is still young, fragile, and requires a lot more experimentation. Dismissing it because of Western classroom failures would be a disservice. ➡️ Where I agree: the US classroom As a parent in a well-resourced American school district, I agree with parts of the article. My children do math on IXL and Khan Academy, learn English on Lexia, read books on Epic. And I often ask: why? Class sizes are manageable. Teachers are highly competent. The human version of these activities - reading together, solving problems collaboratively, learning from peers - would be far richer. Here, EdTech actually displaces a strength. On this point, The Economist is right. But the diagnosis should not become a global verdict. ➡️ A more honest conclusion The same word - EdTech - is used to describe both unnecessary software in teacher-rich classrooms and carefully designed tools in teacher-scarce ones. Conflating them is analytically sloppy and ethically irresponsible. When The Economist declares "EdTech mostly useless" without this distinction, the price is paid by educators in under-resourced systems trying to innovate responsibly, governments deciding where to invest scarce funds, and children who might access quality instruction for the first time. #Edtech

I taught the IB curriculum for 15 years and always respected the core philosophies such as the Learner Profile and the Approaches to Learning (ATL). ATL skills require students to transfer knowledge across subjects, and the idea of embedding AI literacy within ATL makes sense. AI is not a subject; it is a tool and a way to learn. The skills we build around it have to transfer into the learning. When students are asked to question an AI output, compare it to another source, revise it, or defend why they kept certain parts of the output and rejected others, they are doing cognitive work. Teachers are not just covering content; they are adding productive struggle and embedding thinking through ATL skills. These skills show up everywhere, from communication skills that reinforce learning and often lead to “aha” moments, to explaining reasoning in plain language. Through ATL skills, student insight becomes visible. And metacognition is key! It was hard teaching and kids sometimes got tired of always reflecting on their work, but they got really good at it. (And I am often told that I am very meta.) And now we are asking students to consider what changed in their thinking and how they worked through difficult scenarios while using AI. Metacognition is the way! But the IB goes a step further by building in a culture of integrity already through the Learner Profile. We always had posters hanging in every classroom and hallway in the schools I taught at. And when you consistently name characteristics such as principled, reflective, and thinker throughout the K–12 years, students are reminded daily of the expectations we want them to achieve. We are reinforcing that "how they learn" matters as much as "what" they produce. Productive struggle and integrity in learning are not separated from the curriculum for IB schools. And the IB philosophy serves as a strong guide for schools looking to develop their AI literacy within the curriculum intentionally. https://lnkd.in/eAD9JkpD The International Educator (TIE)

From Individual Students to Individuating Learners: Transdialogism and Simondon’s Individuation For Gilbert Simondon, individuation describes an ongoing, metastable process that never resolves into a closed unit or being. What we call an individual is a temporary stabilization within that process, a momentary configuration rather than an ontological starting point. Individuation remains productive because a portion of the preindividual stays active. If that potential were fully discharged, individuation would break down and activity would stall. Educational theory rarely treats this seriously. The dominant figure of the “individual student” presumes a bounded learner who precedes learning. Within this picture, learning becomes interior accumulation, assessment becomes measurement, and personalization becomes optimization. Because the learner is treated as already individuated, individuation itself disappears from educational thought. If individuation is ongoing, it does not occur inside a learner. It occurs through exchanges. In transdialogic contexts, individuation relocates from subject to relation. It distributes across language, interfaces, institutional constraints, technical systems, and interpretive responses. What individuates is a temporary configuration of relations rather than a person understood as a stable unit. Learning outcomes therefore do not belong to learners as possessions. The learner appears provisionally within the exchange rather than standing outside it as a container of effects. This matters because, in transdialogic exchanges, inquiry often narrows early, as the exchange organizes itself around a particular way of posing the problem before alternative framings develop. Early framings determine which distinctions matter, which abstractions organize the exchange, and which interpretations guide what follows. Exchange-based individuation does not reliably expand capacity. When an exchange stabilizes too quickly around a familiar orientation, alternative paths drop out. Responses may appear adequate because they align with an early framing, not because the exchange supported a wider range of possibilities. Individuation continues, but it consolidates a limited configuration instead of extending what the exchange could support. If individuation occurs through exchange, mis-individuation does as well. Responsibility therefore shifts away from learners and toward the architectures of exchange themselves: their design, governance, timing, and modulation. Education, under these conditions, no longer optimizes individuals presumed to exist in advance. It conditions environments of individuation.