Written after reading Vishal R. Choraliya’s essay “Screens are not teachers” in The Hindu — by a founder who builds the very kind of technology the piece warns about, and who thinks the warning is correct.

This week I opened the EducationPlus section of The Hindu and read a piece that, by every reasonable expectation, I should have wanted to argue with. It is titled “Screens are not teachers,” written by Vishal R. Choraliya, an assistant professor at Christ University, Bengaluru. Its argument is that the large-scale push of digital technology into classrooms is harming children, and that India is sprinting into the same mistake the West is now quietly walking back from.

I build an ed-tech company. Karkei puts an augmented-reality layer between a child and a concept. By the easy reading, I am exactly the kind of person who should be defensive about an essay like this.

I am not. I think it is one of the most important things written about Indian education this year. And I want to explain why a founder who builds screens into learning agrees with almost every word of an article that says screens are not teachers — because the agreement is the whole point.

First, the evidence is real. Take it seriously.

In January 2025, Dr. Jared Cooney Horvath, a neuroscientist and education researcher, testified before the United States Senate Committee on Commerce, Science, and Transportation. His central claim is stark: Generation Z is the first cohort in modern history to score lower than the generation before it across attention, memory, literacy, numeracy, and even IQ — despite spending more years in formal schooling than any generation before them.

This is not a vibe. It is a measurement. Horvath correlated classroom-technology adoption across roughly 80 countries with cognitive-performance data and found a consistent pattern: once one-to-one digital devices became standard — a shift that tracks to around 2010 in Western schools — outcomes deteriorated. In his testimony he noted that students using computers for about five hours a day for learning scored more than two-thirds of a standard deviation below peers who rarely or never used classroom technology. Two-thirds of a standard deviation is not a rounding error. It is the difference between a child who is keeping up and a child who is being left behind.

The international data points the same way. PISA 2022, drawing on 81 countries, found that modest device use inside structured learning produced small gains, while heavy screen use correlated with dramatically lower scores. I want you to hold onto that sentence, because it is the hinge on which this entire debate turns, and almost everyone — on both sides — skips it.

The fine print everyone skips

The headline says “screens are not teachers.” The data underneath says something more precise: passive, high-volume, unstructured screen time harms learning, while modest, structured, purposeful technology use does not — and can even help.

This is not me softening the finding. It is the finding. The PISA result and Horvath’s own testimony both endorse selectivity, not abolition. Choraliya says it plainly in The Hindu: the evidence calls for selectivity, not blanket acceleration. The villain in the data is not the device. The villain is the default — a child handed a glowing rectangle, alone, for hours, to absorb content passively while the hard, human work of teaching is quietly outsourced to an app.

That default is real and it is dangerous and India is building it at scale right now. NEP 2020 places digital technology near the centre of its vision. Coaching apps and AI tutors are filling, at enormous speed, the space the state has long failed to occupy. The framing of technology as an equaliser is seductive precisely because it lets governments sidestep the harder work of building human educational capacity — trained teachers, functional schools, real accountability. A screen is cheaper than a teacher. That is the entire problem, compressed into one sentence.

So when the article warns India, it is right. But the warning is not “technology in learning is bad.” The warning is “passive consumption dressed up as education is bad.” Those are very different claims, and the difference is where I have spent the last three years of my life.

There are two kinds of screen

Imagine two children, each holding a tablet.

The first child is watching. A video plays. A voice explains. Occasionally a multiple-choice question appears and the child taps an answer to make the video continue. The child is physically still, cognitively passive, and emotionally absent. The brain treats this stream the way it treats most things that arrive without effort: as low-priority information to be cleared. This is the screen Horvath measured. This is the screen that is making a generation forget. It deserves every word of the warning.

The second child is also holding a screen. But her hands are moving. The gesture she makes with her fingers rotates a three-dimensional human heart in the air in front of her. She triggers the pumping cycle herself. She follows the path of oxygenated blood because she chose to chase it. She is not watching someone else learn. She is doing the learning with her own body, her own choices, her own curiosity driving every frame. The screen here is not the teacher. It is the instrument — the way a microscope is an instrument, or a piano, or a potter’s wheel.

Neuroscience does not treat these two children the same, and neither should policy. The first child is in passive intake; almost none of the conditions for durable memory are present. The second child is generating, moving, and curious — the three states under which the brain actually consolidates what it encounters into long-term memory. Same hardware. Opposite outcomes. The screen is not the variable. What the screen asks the child to do is the variable.

The tragedy of the current ed-tech wave — in the West and now in India — is that almost all of it builds the first kind of screen and calls it the second.

What Karkei actually is — and is not

I started Karkei because I have three daughters and I watched them study for hours and remember almost nothing, and I became convinced the failure was one of design, not of children. Karkei is an augmented-reality experiential-learning system. We deliberately built it on a no-headset architecture, because the moment learning requires a ₹30,000 device, it stops being a tool for Indian classrooms and becomes a toy for wealthy ones — and that inequity is the second half of Choraliya’s argument, which I will come to.

Every experience we build is run through an internal framework we call the SPARK Method — Simulate, Play, Act, React, Know. Each letter is a test we apply before anything ships. Does it simulate something real enough to reach in and touch — a beating heart, an atom, a circuit? Can the child play with it freely, rotating it, opening it, exploring with no wrong moves? Does the model act — come alive and respond on cue, rather than sit there static? Does it let the child react — predict, test, and be answered back, a real feedback loop? And does the lesson finally land so the child knows it in the body, not just the notebook? These five tests are not marketing. They map, almost one-to-one, onto the conditions cognitive science says are required for memory to form. We did not invent the science. We built an engineering discipline around refusing to ship anything that violates it.

Crucially, Karkei is not built to replace the teacher. It is built to give the teacher a more powerful instrument. The article’s deepest point — the one Horvath keeps returning to — is that human cognition evolved to learn from other humans. I believe that completely. A Karkei session without a teacher in the room is a worse session. The technology is there to make the ten minutes a teacher spends on the human heart land harder, stick longer, and reach the child in the back row who normally switches off. The human stays at the centre. The screen serves the human. Reverse that hierarchy and you have built exactly the thing the article warns against.

What we measured in a Chennai classroom

I do not ask anyone to take this on faith, because the whole argument of the article is that ed-tech has been sold on faith and advocacy rather than evidence. So here is evidence, with all its smallness stated honestly.

In April 2026 we ran an early pilot of Karkei’s SPARK-based Heart Module at a Chennai school. Thirteen Grade 10 students. Five teachers, with between 7 and 18 years of experience, observing from the back. One short, gesture-driven AR session, bracketed by handwritten surveys and a diagnostic.

Mean self-reported understanding of how the heart works rose from 3.1 to 4.1 on a five-point scale in a single session. Every student who completed the diagnostic said they wanted more subjects taught this way. All five teachers said they would recommend Karkei to colleagues, rated observed engagement at 4.6 out of 5 against a typical Biology class, and — the signal I keep returning to — every one of them noted that students who normally disengage came back into the room. The student who rated the controls worst, who clearly found the interface frustrating, recorded the largest objective knowledge gain in the room. The content reached her even through friction, because it was concrete, visual, and hers to manipulate.

Thirteen students is not a study. I will be the first to say it, before any critic does. It is a signal, not a proof. Which brings me to the part of the article I want to take most personally.

The standard I am holding Karkei to

Horvath’s sharpest recommendation to the Senate was not “ban screens.” It was this: any publicly funded ed-tech should be required to demonstrate its learning benefit through independent, replicated trials before it is deployed to children. Most of the industry treats that as a threat. I want to treat it as a commitment.

If a technology cannot show, under independent scrutiny, that children using it learn more and remember longer than children who do not, then it has no business in a classroom — and that test applies to Karkei as fully as it applies to anyone else. So this is the standard we are building toward: not testimonials, not engagement metrics that merely prove a thing is fun, but controlled, repeatable evidence of durable learning, published whether it flatters us or not. The day Karkei stops being able to clear that bar is the day Karkei should stop existing. I would rather build inside that constraint than pretend the warning in The Hindu does not apply to me.

The part of the warning India must hear loudest

There is one thread in the article that should keep every Indian founder, policymaker, and parent awake. The historian Eliza Filby, quoted in the piece, notes a quiet divergence in the West: the very families who build and sell these technologies are pulling their own children away from screen-heavy schooling and back toward teacher-led instruction, while the children most exposed to device-led learning are increasingly those in underfunded public schools. The people who know exactly what they built are choosing not to feed it to their own kids.

India is about to import that divergence wholesale. The risk is not that screens flood every Indian classroom equally. The risk is that elite private schools quietly keep their low ratios and their human teachers, while government and low-fee schools — serving the children who can least afford a worse education — get handed cheap tablets loaded with passive content, and we call it equity because everyone now has “access.” Access to the wrong thing is not equity. It is abandonment with a charger.

This is precisely why I refuse to build Karkei as a premium, headset-dependent, high-cost product. If embodied, well-designed, teacher-augmenting learning is only affordable to children who are already advantaged, then I will have built one more instrument of the divergence the article is warning about. The entire point is to put the good kind of screen — the active, structured, human-led kind — into the classrooms that would otherwise be given only the bad kind. Get that wrong and the technology is part of the disease. Get it right and it is part of the answer.

Where I land

Read “Screens are not teachers” if you have not. It is the most clear-eyed thing in the Indian education conversation right now, and I am grateful to Choraliya for writing it and to The Hindu for running it. Then read the original source it draws on — Horvath’s Senate testimony — because the data is more interesting and more useful than the headline.

And then hold the right idea, not the easy one. The easy idea is “technology is ruining learning, pull the plug.” The right idea is harder and more demanding: passive screens are ruining learning, and the answer is to be ruthlessly selective about technology, to keep the human teacher at the centre, and to demand independent evidence before anything reaches a child.

Screens are not teachers. They never were. The good ones were never trying to be. They are instruments in a teacher’s hands — and the whole job, the only job worth doing in this field, is to make sure that the screen serves the human, and not the other way around.

That is the line the article draws. It is the line Karkei is built on. And it is the line India cannot afford to get wrong.

— Elango Raghupathy
Founder, Karkei · Chennai · elango@karkei.com

Karkei is a Chennai-based company building SPARK-compliant, no-headset experiential-learning systems for Indian schools. Our modules are currently in classroom pilots across Tamil Nadu. If you teach, lead a school, or are a parent who wants to know more, write to us at vanakkam@karkei.in.

References & further reading

  1. Vishal R. Choraliya, “Screens are not teachers: What India can learn from the West’s ed-tech reckoning,” The Hindu, EducationPlus (“Get the Edge” column), June 2026. The essay that prompted this piece.
  2. Dr. Jared Cooney Horvath, Written Testimony before the U.S. Senate Committee on Commerce, Science, and Transportation, 15 January 2025. commerce.senate.gov — the original source, including the 80-country analysis and the recommendation for independent, replicated trials.
  3. OECD, PISA 2022 results on digital device use and student performance across 81 countries — the finding that modest, structured use yields small gains while heavy use correlates with sharply lower scores.
  4. Dr. Eliza Filby, historian of generational change, on the widening divergence between device-light elite schooling and device-heavy public schooling (quoted in The Hindu, above).
  5. Selected coverage of the Horvath testimony: Fortune and Chalkbeat (which also weighs the counter-evidence).