A Human Being is a Possibility: Rethinking School Through One Learner at a Time

Return to Knowledge Garden
A Human Being is a Possibility: Rethinking School Through One Learner at a Time
When I think about the future of Indian education, I keep coming back to a very simple frame:
One learner at a time.
Not one batch.
Not one section.
Not even one “average” student profile.
One actual human being at a time—with their own pace, fears, spark, background, and possibilities.
"A human being is a possibility, not a resource."
– Sadhguru
If we take that sentence seriously, almost everything about how we design learning has to change. Possibilities cannot be mass-produced; they must be recognised, nurtured and accompanied. Earlier pieces in this series have looked at curriculum lag, bridges like SETU for equity, ANKUR as the metaphor of each child’s growth, UTSAH as the spark of joy, Holistic Progress Cards as 360‑degree mirrors, and DARPAN as the reflection between digital and physical worlds.
This article sits at the centre of that constellation. It is about personalisation not as jargon, not as a product, but as a design commitment:
How do we change the world? One learner at a time.
A classroom story: the day the lesson slowed down for one child
Imagine a middle school classroom in a government-aided school.
Forty-two students. One teacher. A chapter on fractions that must be “finished” this week.
At the front, the teacher works quickly through examples: 1/2, 3/4, 5/8. Some students nod along. Others copy mechanically. On the third bench from the back, Riya’s eyes glaze over. She has memorised procedures in the past, scored decently, and promptly forgotten them.
Today, something different happened.
After the board work, instead of moving straight to more problems, the teacher pulls out a small set of shared devices. She pairs students, one device for two. The app they open does something very specific:
It starts with two very simple tasks: fractions that almost everyone can solve.
If a pair answers quickly and correctly, it gently increases difficulty.
If they hesitate or answer incorrectly, it steps back to more concrete tasks, adding visual models and hints.
Within minutes, the system figures out that Riya can recognise halves and quarters, but gets confused when denominators differ. It stops throwing random problems and stays with that precise idea: “How do you make unlike denominators talk to each other?”
Her partner moves faster through the levels and hits challenge problems like: “Which is larger, 3/5 or 4/7? Explain why.” He is stretched, not bored.
At the end of the session, the teacher looks at her dashboard. She doesn’t see a list of marks. She sees concept heatmaps: which micro-skills clicked for which students, who needs manipulatives tomorrow, who is ready for real-world word problems.
Later, in a quiet moment, she kneels next to Riya’s bench.
“I noticed the app slowed down for you at common denominators,” she says. “Tomorrow we’ll try a different way, with rotis and paper strips. Will you help me?”
Riya nods, surprised. For the first time in a long time, she feels that the class has noticed her.
“One child, one teacher, one book, one pen can change the world.” – Malala Yousafzai
“One child, one teacher, one book, one pen can change the world.” – Malala Yousafzai
Personalisation is what happens when that sentence becomes more than a poster: when a teacher, supported by AI, can see one child at a time and respond with intention.
What personalisation is and what it is not
Personalised learning can mean many things. To move beyond buzzwords, it helps to clear two persistent misunderstandings that thinkers like Dwayne Harapnuik and others call out.
Myth 1: Personalisation = more software
Too often, schools assume that buying a new platform equals personalisation. Harapnuik calls this the search for a quick fix: the belief that a new model or app will solve deep learning problems without changing mindsets or environments. History shows that technology, when “bolted on”, is often oversold and underused it does little more than digitise old habits.
In reality, platforms can:
Diagnose where a learner stands.
Adjust difficulty and sequence.
Collect fine-grained data over time.
But they cannot:
Build trust with a child.
Understand family context and local culture.
Design authentic tasks that matter.
Help a learner make meaning of their own journey.
“Education is not the filling of a pail, but the lighting of a fire.” – William Butler Yeats
“Education is not the filling of a pail, but the lighting of a fire.” – William Butler Yeats
AI can help with the pail organising content, sequencing questions but only a human teacher lights the fire. Real personalisation starts with the fire.
Harapnuik’s work emphasises significant learning environments where learners have Choice, Ownership and Voice through Authentic tasks (COVA in a CSLE). AI can support such environments; it cannot create them on its own.
Myth 2: Personalisation = everyone isolated on a device
The second misconception is that personalisation fragments the class into isolated silos: each student on a screen, learning alone.
In fact, the most powerful personalised environments whether in schools or workplaces blend:
Individual paths for key competencies (like decoding or number sense).
Collaborative projects tackling real problems.
Shared standards and goals, so everyone is ultimately aiming at common competencies.
Personalisation is not about letting students drift apart. It is about ensuring each learner is appropriately challenged and supported, instead of being perpetually lost or perpetually un-stretched.
One learner at a time: the mindset behind the method
Harapnuik’s phrase “changing the world one learner at a time” is not a slogan; it is a discipline. It rests on three convictions:
Learning is a process, not an event.
There is no once-and-for-all “fix”. Leaders and teachers must commit to ongoing cycles of analysis, evaluation and creation in their environments.
Quick fixes are lower-order thinking.
When we only search for models to “apply”, we operate at the apply level of Bloom’s taxonomy adequate for routine tasks but insufficient for the complex demands of a digital world.
Deeper personalisation requires higher-order thinking for teachers and students.
To personalise meaningfully, educators must analyse their learners’ contexts, evaluate what is or isn’t working, and create new designs. Students must move beyond rote to analysing, evaluating, and creating in their own work.
“The goal of education is not to increase the amount of knowledge but to create the possibilities for a child to invent and discover.” – Jean Piaget
“Possibilities” is the key word. NEP 2020 shares this insistence that we move beyond rote, one-size-fits-all learning to competency-based, higher-order and experiential forms of education. Personalisation done well is simply the practice of that philosophy, one child at a time.
NEP 2020 and personalisation: reading the policy through a learner’s eyes
When we read NEP 2020 with “one learner at a time” as our lens, a clear thread emerges.
From coverage to competency
NEP 2020 calls for a decisive shift:
From syllabus coverage to mastery of core competencies.
From memorisation to understanding, application and critical thinking.
From rare high-stakes exams to continuous, formative assessment that actually supports learning.
This translates into:
Progress based on what students can do, not just what they have “completed”.
Assessments that capture a 360-degree picture hence the Holistic Progress Card work through PARAKH rather than only right-or-wrong answers.
Teachers using assessment for and as learning: adapting instruction and helping students self-reflect.
AI-enabled personalisation depends on exactly this kind of granular competency information: what each learner understands today and what they are ready for next.
Flexibility and choice as personalisation mechanisms
NEP brings in:
Subject flexibility in secondary school.
Multidisciplinary combinations.
Exposure to vocational skills and local knowledge.
Policy commentary on personalisation under NEP highlights:
The need for learner profiles that track strengths, interests and goals.
The centrality of student agency in setting goals and reflecting on growth.
The importance of teacher training in differentiated and personalised instruction.
Personalised learning paths are how these abstract commitments show up on the ground: one learner chooses a STEM-heavy route, another blends arts and computing; both work at appropriate levels on shared competencies like communication and problem solving.
Technology and AI as enablers not replacements
NEP 2020 sees technology and AI as:
Tools for adaptive learning that adjusts to pace and level.
Engines for real-time analytics on student progress.
Supports AI-based tutoring and guidance beyond the classroom.
“The country is not made of bricks, it is made of consciousness.
Only when people are enlightened, the country be enlightened.”
– Rabindranath Tagore
If we accept Tagore’s insight, then personalised, competency-based learning supported by AI and rooted in strong teacher–student relationships is not a luxury. It is how we build the consciousness that NEP 2020 dreams of, one learner at a time.
Putting the person back into personalised learning
In a recorded conversation on personalised learning, Harapnuik and colleagues warn against confusing adaptive software with personalisation itself. They make a simple but crucial point:
Software is powerful for practice and progression.
The personal dimension, relationship, mentoring, guidance comes from the teacher.
“The supreme art of the teacher to awaken joy in creative expression and knowledge.” – Albert Einstein
AI can sift data faster than any human. It can recommend resources, detect patterns, and suggest next steps. But only a teacher can look a child in the eye and say, “I see you. I believe you can do this. Let’s try again, together.”
This is deeply aligned with NEP’s view of teachers as:
Facilitators and mentors, not mere content transmitters.
Observers and guides who use formative assessment wisely.
Key anchors of socio-emotional development in schools.
In this model, AI and teacher play complementary roles:
AI handles the routine diagnostics and routing.
The teacher handles meaning-making, motivation, and moral guidance.
Personalisation, then, is not about machines replacing humans. It is about machines doing what they do best so that humans are freer to do what only they can do.
Designing for concept mastery, not mugging
At the centre of all this lies a deceptively simple shift: from mugging to mastery.
Traditional systems:
Present a chapter.
Drill students with similar questions.
Conduct a test.
Move on, regardless of actual understanding.
Personalised, competency-based systems:
Break concepts into smaller, clear learning targets.
Use formative checks to track each learner’s status against those targets.
Provide targeted support or additional challenges based on that status.
Allow flexible time so different learners can reach the same standard at different speeds.
Earlier in this series, I wrote about Item Response Theory as a technical backbone for precisely this kind of adaptive, competency-based assessment. AI builds on such models to:
Recommend the right question at the right difficulty.
Detect when a student is guessing.
Suggest varied contexts so that only conceptual understanding (not pattern memorisation) succeeds.
In such a world, the system is designed so that mugging is insufficient. Students are sometimes compelled to think, apply, explain, create.
“The aim of education is knowledge, not of facts, but of values.” – William Ralph Inge
“The aim of education is knowledge, not of facts, but of values.” – William Ralph Inge
If personalisation only helps students cram facts faster, we will have failed. One learner at a time must also mean one set of values, one conscience, one capacity to use knowledge responsibly at a time.
Stories from the ground: personalisation in Indian classrooms
To keep this rooted, let’s return to the ground composite stories based on emerging practice.
Story 1: The reading ladder
In a Class 3 classroom with wide reading levels, the teacher replaces whole-class choral reading with a blended routine:
Once a week, students read short texts of their choice on shared devices, folktales, simple science, biographies.
An AI-enabled tool tracks reading speed, accuracy, and hesitation points using speech recognition.
A profile emerges: which sound patterns confuse whom, who decodes well but struggles with meaning, who is ready for more complex syntax.
Her dashboard clusters learners:
Group A: struggling with specific phonics patterns.
Group B: decoding fine, comprehension weak.
Group C: ready for inferential questions.
She then re-designs instruction:
Group A gets targeted phonics games and concrete activities.
Group B works through read–retell–draw cycles to deepen meaning.
Group C engages in book clubs and mini-projects.
For the child who used to hide in the back, guessing words, the change is profound: the system is not branding them “weak”; it is meeting them where they are and walking with them.
Story 2: The quiet coder
In a Class 9 section, Arjun rarely speaks. He does his work, gets average marks, blends in.
When the school introduces a simple AI literacy and coding unit, the teacher designs it with personalisation in mind:
Students choose a problem they care about: water wastage, canteen queues, bus timings.
They use block-based tools to build small prototypes.
An embedded AI assistant suggests next steps: “Try a loop here”, “Track another variable”.
Arjun chooses to model electricity usage at home. After school, he quietly experiments, guided by the AI’s hints. The teacher sees his project logs and realises he is attempting more complex structures than most.
Instead of just grading the final submission, she:
Invites him to share his model in a small group.
Connects his work to upcoming physics concepts.
Encourages him to record a screen-cast, narrating his logic.
Suddenly, Arjun’s identity is shifting from anonymous worker to “the one who built that model”. AI gave him a safe, low-judgement space to tinker. The teacher turned that into visibility, affirmation, and a path forward.
Story 3: The SETU plan in a rural cluster
In a rural block, assessments reveal large foundational gaps. The cluster adopts a modest personalised strategy:
Simple offline tests are periodically digitised at the block office.
An AI model identifies conceptual patterns where exactly students are stuck.
Resource teachers design SETU plans: focused cycles for small groups, built around local contexts rather than generic worksheets.
One teacher notices that several students can perform division algorithmically but cannot explain what the operation means. The AI has flagged this pattern: high scores on procedure, low on word problems.
Her intervention:
A simulated local market in class, with grains, coins, and land plots.
Story problems using their own families’ livelihoods.
Reflection prompts like, “Where do we see this kind of sharing at home?”
She records their explanations as short audio clips. These become part of each child’s ANKUR portfolio, evidence of conceptual understanding sprouting over time and later feeding into Holistic Progress Cards.
Connecting the dots: DARPAN, ANKUR, SETU, UTSAH
Across this series DARPAN, SETU, ANKUR, UTSAH, AI as equaliser, Holistic Progress Cards a pattern is emerging.
DARPAN (the mirror) is about aligning experiences across digital and physical spaces. In personalisation, DARPAN means AI insights from digital practice inform classroom projects, and lived experiences feed back into digital profiles.
ANKUR (the sprout) is about tracking small but meaningful growth in identity and competence. Personalised pathways and AI-supported portfolios make ANKUR visible in ways old report cards never could.
SETU (the bridge) stands for connecting learners across divides. Adaptive tools in local languages, offline modes, and simple dashboards help ensure that children in government and low-fee schools are not last in line.
UTSAH (the spark) is the joy at the heart of learning. AI can either flatten UTSAH with sterile drills, or amplify it with stories, challenges, and choice. The design choice is ours.
“How do we change the world? One random act of kindness at a time.” – Morgan Freeman
In our context, a “random act of kindness” might be as simple as a teacher using AI data to notice a quiet learner, redesign a task, or recognise a hidden strength. Over time, those acts multiplied across schools change more than individual lives. They change culture.
Cost–benefit: can we afford one-learner-at-a-time designs?
At system level, it is fair to ask: is “one learner at a time” realistic in India?
The invisible cost of one-size-fits-all
Learning poverty, disengagement, and skills gaps are not free:
Students reach secondary school lacking basics.
Families spend heavily on private tutoring.
Employers face graduates with certificates but not competencies.
These are long-term social and economic costs that we rarely tally alongside budgets for devices or platforms.
The economics of AI-supported personalisation
Studies of AI-driven adaptive learning and assessment in Indian contexts show that:
Once infrastructure and content are in place, the marginal cost per student can be low.
AI can significantly reduce manual tasks, making teacher time more impactful.
Early, targeted interventions can prevent expensive remedial programs later in schooling.
Viewed against the alternative, investments in AI-supported personalisation look less like luxuries and more like strategic necessities for NEP’s goals.
“A human being is a possibility, not a resource.”
If we design as if learners are only economic resources, we will count costs narrowly. If we design as if they are possibilities, we will invest differently.
What each stakeholder can do starting now
Because this article speaks to multiple audiences, here are concrete invitations.
For school administrators and principals
Make “one learner at a time” explicit in your school vision and connect it to NEP 2020’s language on personalisation, flexibility, and holistic development.
Invest first in teacher design capacity choice-rich tasks, formative assessment, basic data use before complex platforms.
Start with small pilots: one grade, one subject, one AI-supported workflow. Study impact on understanding and engagement.
Align report cards towards Holistic Progress Cards so that what you measure matches what you value.
For teachers and educational leaders
Shift the central question from “How do I finish the syllabus?” to “How will each learner make meaning?”
Use AI as a lens, not a leash: let it surface patterns, but keep judgment human.
Normalise self and peer assessment so students gradually own their trajectories.
Build relational rituals, goal-setting chats, feedback circles into personalised routines.
For policymakers and government officials
Strengthen NEP-aligned guidance on personalisation, competency-based progression, and ethical AI use in classrooms.
Ensure AI deployments prioritise government and low-fee schools, with multilingual, low-bandwidth designs.
Bake privacy, consent, and algorithmic fairness into regulations from the outset, especially when data feeds into high-stakes decisions.
For parents and education advocates
Change the home conversation from “Kitne marks?” to “Kya samjha?” and “Kya naya seekha?”.
Celebrate perseverance, empathy, and curiosity alongside scores.
Ask schools how technology helps your child learn better, not just faster.
For academic researchers
Study how AI and pedagogy interact in Indian classrooms, with disaggregated data for different social groups.
Contribute theory from Indian contexts: what “one learner at a time” looks like in real government schools, not just in idealised settings.
“The country is not made of bricks, it is made of consciousness.”
Personalisation at scale is ultimately a consciousness project: systems becoming aware of individuals, and individuals becoming aware of themselves as learners.
Closing reflection: the progress we cannot see on a marksheet
At the end of Class 10, a student holds a document that attempts to summarise school. In the old model, that document was mostly a tally of marks. In the emerging model, especially with Holistic Progress Cards and AI-augmented portfolios, it can become something else:
A mirror, reflecting back growth in understanding, character and capability.
A window, through which others see more than a percentage.
A map, pointing towards futures that align with real strengths and aspirations.
AI, used wisely, does not replace this meaning-making. It helps organise the evidence, the quiz results, project artefacts, reflections, feedback so that teachers and students can see patterns in time to act.
“A human being is a possibility, not a resource.”
“The goal of education is not to increase the amount of knowledge but to create the possibilities for a child to invent and discover.”
Between those two sentences lies the work of this decade in Indian education. NEP 2020 gives us the policy language. SETU, ANKUR, DARPAN, UTSAH, AI as equaliser, and Holistic Progress Cards give us conceptual tools.
“One learner at a time” is how all of that becomes real.
Not someday, in some ideal system.
But today, in this classroom, with this child, guided by this teacher, supported by this quiet AI in the background turning potential into progress, and progress into possibility.