Report: AI for children should not bypass human validation: Vineet Nayar, Founder, Sampark Foundation

Latest updates suggest that, Vineet Nayar is the Founder-Chairman of Sampark Foundation and the former CEO of HCL Technologies.During his tenure, Vineet led the transformation of HCL Technologies into one of India’s fastest-growing global information technology service companies. New information indicates that, Vineet stepped down from HCL Technologies to focus on Sampark Foundation, which he co-founded with his wife Anupama Nayar., His management practices at HCL Technologies became case studies at bothHarvard Business School and London Business School.In 2013 The foundation works to drive ‘large-scale transformation in primary education through frugal innovation’.Vineet holds a degree in mechanical engineering from GB Pant University of Agriculture and Technology and a management degree from XLRI Jamshedpur. He is also the author of the management bestseller, Employees First, Customer Second: Turning Conventional Management Upside Down (Harvard Business Press). Vineet spoke to indianexpress.com on the work of the Sampark Foundation, the challenges of transforming education in India through technology, and how AI could transform the education sector in rural India. In a recent development, Edited excerpts: Venkatesh Kannaiah: Tell us about your journey from HCL Technologies to the Sampark Foundation. Vineet Nayar: My journey from HCL to Sampark Foundation was not about moving from business to philanthropy. Latest updates suggest that, It was about moving from one large transformation challenge to another. Sampark Foundation today reaches 20 million children across two lakh government schools through eight state government partnerships.Story continues below this ad It is not as if we picked up education as a possible intervention in the first instance. including health, rural development, agriculture, water management, and education, and we introduced projects in each one of them., We identified five areas We discovered that the education sector offered the possibility of impact at scale. and my mother was a government school teacher for 30 years., My wife is a special education teacher from the Spastic Society That was of great support. We experimented, learned, and eventually honed in on education. It was an interaction with my mother which made me realise the need to find one’s true meaning in life. Latest updates suggest that, And true meaning, she stated, comes only when you support others achieve their goals. when I was 50 years old, that I realised the need to change my path so that I would still have the energy to execute another transformation.Story continues below this ad Venkatesh Kannaiah: What were the initial challenges?, That was the moment of change around 2011-2012 Vineet Nayar: When I entered the education sector, I saw that millions of children were in school, but learning was not happening. The system had become dependent on compliance instead of inspiration. Teachers were overloaded. New information indicates that, Classrooms were overcrowded. According to recent reports, Solutions were designed in boardrooms, not for the realities of rural India. I realised that if we wanted transformation at scale, we had to combine three things: technology, simplicity, and the power of human connection. That became the foundation of Sampark. In a recent development, What excites me most is that Sampark is not a charity, distributing products. We are building scalable learning systems for Bharat. New information indicates that, Venkatesh Kannaiah: Tell us about your learnings from HCL that you have brought forward to the Sampark Foundation. And what is it that you had to unlearn? In a recent development, Vineet Nayar: The world often behaves as if machines are replacing humans. In a recent development, My experience has been the opposite. Machines amplify humans. Weak systems become weaker with technology. Strong human systems become transformational.Story continues below this ad At HCL, I also learned that large systems change only when frontline people are empowered. In education, the frontline is the teacher. Yet most education reform treats teachers as the problem instead of the solution. So we asked ourselves a different question: how do we make teachers successful? How do we reduce complexity inside classrooms? According to recent reports, The second lesson was about scale. New information indicates that, India needs solutions that can work across millions of children. In a recent development, affordability, and simplicity at its core., Every Sampark innovation is designed with scale The third lesson was around feedback loops. accountability. and Technology becomes powerful when it creates visibility New information indicates that, Without data, systems drift. systems improve.Story continues below this ad On what I needed to unlearn, the amount of resources I had at HCL Technologies was almost unlimited, and I could pick up the phone and speak to anyone in the world., With data All of that collapsed overnight. So I had to build the organisation all over again, this time with people who were interested in villages and rural education, which is a much smaller pool. People no longer picked up your calls just because you held a particular chair or title. In a recent development, And yes, we had resources, but not billions of dollars to take unlimited bets. So that was one major level of change. The second level of change was emotional and cultural. At HCL, it was a for-profit initiative. but here the morale of the employees is at a level I had never experienced before., The enthusiasm and motivation were certainly there As a result, my own motivation, morale, and sense of satisfaction became far higher. In a recent development, Venkatesh Kannaiah: Where does tech fit in the Sampark Foundation’s vision? New information indicates that, Vineet Nayar: Technology sits at the centre of Sampark’s work, but not in the way Silicon Valley imagines it. In a recent development, We are not obsessed with shiny technology. internet, training, and ideal classroom conditions., We are for solving real problems at scale.Story continues below this ad Most educational technology fails in rural India because it assumes infrastructure Rural India has none of these consistently. Venkatesh Kannaiah: Tell us more about one of your innovations, Sampark TV, and how it works on the ground? Vineet Nayar: When you enter a classroom, you will see all the content aligned with the state textbooks, almost in a Netflix-style interface. According to recent reports, The entire system is driven by an Android device connected to a smart board. It does not require internet connectivity. pedagogy is based on the idea that there is a structured way of teaching every microconcept correctly., Now You will see lesson strategies, animated videos, and there will be gamified assessments that teachers conduct in the classroom. Latest updates suggest that, sends classroom data back to us. and There is also an AI engine inside Sampark TV that captures We process it through the AI engine and share daily insights with government agencies.Story continues below this ad Normally, we do not know what is happening inside classrooms. while also allowing us to understand what is actually happening in the classroom., What Sampark TV has done is make teaching easier and learning more engaging According to recent reports, we are able to extract classroom-level data and provide it to state administrators., At the same time In a recent development, They can see which teachers are teaching based on plan and which are not. The teacher does not have to independently explain every microconcept. She becomes an enabler of learning rather than the sole source of instruction. The multimedia content helps deliver the concept. I am completely against devices in the hands of children, and I can cite research to support that view. I believe the device should remain in the hands of the teacher. So what you will see is one large Android screen in the classroom, fully controlled by the teacher. No personal devices in the hands of students. It is currently active in 86,000 classrooms.Story continues below this ad Venkatesh Kannaiah: Tell us about Sampark AI and how it works. In a recent development, Vineet Nayar: Sampark AI helps us design the optimal learning path for every microconcept. how the children are responding. and The Sampark AI engine inside the Android device captures data on what the teacher is teaching In a recent development, which then feeds insights to government administrators through dashboards., That data flows into our AI engine They can see which teachers are progressing based on plan and which are not. Sampark AI helps us assess children periodically to understand whether learning is actually happening. New information indicates that, Venkatesh Kannaiah: How does the system track individual student-level performance? Vineet Nayar: So the first level of tracking happens at the classroom and community level. We visually identify the responses and understand what percentage of children selected the correct answer. In a recent development, That completes the classroom-level assessment. When a Sampark staff member visits the classroom, they take photographs of the children’s workbook pages. Latest updates suggest that, Then AI converts those photographs into correct and incorrect responses, and that gives us assessment data for each child individually. Venkatesh Kannaiah: Tell us about the Harvard Business case studies on Sampark Foundation. Vineet Nayar: The Harvard case study on Sampark explores a fascinating challenge: can large-scale social transformation happen with the same rigour, innovation, and accountability expected in the corporate world? New information indicates that, behavioural insights, operational discipline, and government partnerships., The cases examine how Sampark built scalable education systems by combining technology Sampark Foundation also features among the 10 cases in the book Genius at Scale by Harvard professor Linda Hill. Venkatesh Kannaiah: Education in rural India. How much of it is a tech problem? Vineet Nayar: Rural education is not fundamentally a technology problem. New information indicates that, It is a systems and human capability problem. Technology cannot replace motivation, teacher ownership, leadership, or community participation. But technology can dramatically amplify outcomes when used correctly. The mistake many people make is believing that distributing devices equals educational reform. New information indicates that, It does not. In a recent development, A tablet without pedagogy is just hardware. India must be careful not to copy Western models blindly. Our solutions must be built for multilingual classrooms, low infrastructure environments, and first-generation learners. Venkatesh Kannaiah: Tell us about three tech solutions that have not worked in an Indian educational context. Vineet Nayar: The first failure has been hardware-led thinking. India spent years believing that tablets and computers alone would improve learning. They did not. The second failure has been internet-dependent models. According to recent reports, Rural India cannot rely on uninterrupted connectivity for learning. The third failure has been importing urban edtech models into rural government schools. Most of these products were designed for affluent, English-speaking environments and collapsed when exposed to the realities of Bharat. Venkatesh Kannaiah: Tell us about the challenges the education ecosystem would face from new tech interventions? Vineet Nayar: Let’s talk about AI companies first. AI companies want subscribers. they offer free services, but eventually they begin charging for them., Initially In a recent development, Now, despite knowing that AI can fudge facts and hallucinate, we are still putting these tools into the hands of children through subscription-based models. We do not know what kind of history, geography, or understanding a child may absorb over the next two or three years. because their valuations rise when they can claim a billion or two billion users., Companies push AI into the hands of children primarily to drive subscriber growth many state governments in India are adopting them., And because these products are initially free Latest updates suggest that, Now they are training their AI engines using our data. there may no longer be viable alternatives., Once people become dependent on these systems there were operating systems from companies like Wipro, Infosys, and HCL Technologies., I often give this example: when I first entered the technology industry Then Microsoft entered the market, and its operating system spread rapidly because pirated copies were freely available. Latest updates suggest that, Over time, Indian operating systems disappeared. I believe something similar could happen here. build dependence, and later move to subscription models., These companies will learn from our data in my view, are among the biggest contributors to this problem., The edtech companies are now pushing the same approach into rural education. and They built subscription-driven models for urban children In our villages and towns, we have multi-grade classrooms. Classes 1, 2, and 3 often sit together. Latest updates suggest that, We have multiple languages. Attendance may only be 120-130 days a year. And teachers spend only around 19% of their time teaching, while 81% goes into administrative work. According to recent reports, In this context, the tool should be in the hands of the teacher, not the child. That is why I am strongly against putting iPads or similar devices directly into the hands of children. Latest updates suggest that, I am strongly against edtech models that isolate the child with a screen. According to recent reports, It reminds me of the “Free Basics” debate, when Facebook proposed free internet access. In a recent development, the Indian government did not approve of the same., Thankfully According to recent reports, policymakers should apply the same level of scrutiny to these education models as well., In my opinion Venkatesh Kannaiah: Do you work with startups? Tell us about some opportunities that you see for social impact startups in the edtech sector. Latest updates suggest that, many of them using AI and advanced technology to solve relevant business and social problems., Vineet Nayar: I mentor 19 startups Some are solving for teacher productivity and learning personalisation. Others are improving employee assessments, operational efficiency, healthcare delivery, financial models, drug discovery, or enhancing human capability. In education, the biggest opportunities are in foundational learning, multilingual AI systems, adaptive learning, teacher enablement, and low-cost, scalable infrastructure. According to recent reports, I personally commit a certain number of hours every month to mentoring these 19 startups. I do not invest in them, nor do I charge a fee. According to recent reports, The first question when startups approach me is: What is the use case you are trying to solve? Latest updates suggest that, Secondly, is the benefit actually reaching the end beneficiary? And thirdly, is there a meaningful technology embedded in the solution? New information indicates that, If a startup can answer those three questions convincingly, I am willing to mentor them. It does not matter whether they are for-profit or not-for-profit. Venkatesh Kannaiah: How will AI impact the work of the Sampark Foundation? Vineet Nayar: AI can be a powerful tool in helping us understand how microcompetency should be taught. New information indicates that, It can support define the right sequence and the right method of teaching a microconcept. New information indicates that, it can support nudge the teacher., Secondly Suppose I am teaching a concept, and the children are not responding well. especially multi-grade classrooms, you may technically be teaching Grade 3 while several children are still functioning at a Grade 2 level., In many classrooms AI can support identify where learning is breaking down and suggest the next pathway for the teacher. AI can support analyse classroom data and feed insights to administrators and bureaucrats., Third which are not. and It can support identify which schools are teaching based on plan Where AI should not be used is by placing it directly into the hands of children. Latest updates suggest that, AI systems are still not sufficiently human-tested or context-sensitive for unrestricted child use.