Two decades ago, a dangerous idea took hold in some educational technology circles: that teachers were an obstacle to learning, and that well-designed digital content could effectively bypass them altogether. The vision was simple: give a child a screen with flashy animated math and science content, and the ‘middleman’ of the classroom instructor becomes unnecessary.
It was a profound misreading of how education actually works.
The intervening years have demonstrated what should have been obvious from the start: without a teacher, there is no education, only consumption. Technology’s role should not be to replace educators but to empower them, transforming the ‘Sage on the Stage’ into the ‘Guide on the Side’ while preserving the essential human element that makes learning possible.
Fast forward to January 2026. This lesson bears repeating now, as the global conversation around artificial intelligence in education reaches an all-time high. At the World Economic Forum in Davos in January 2026, representatives from Brookings, the Financial Times, and Gallup gathered to examine a new iteration of an old delusion: the promise from the big techs that AI will finally ‘personalise’ education so perfectly that teachers are now truly obsolete.
The consensus from the ‘Tech as a Tool, Not the Teacher’ panel delivered a thunderous wake-up call. Education is not a data transfer; it is a social construct. One cannot download character, empathy, or critical thinking from an LLM. As experts from Canada and the US argued, learning is an interpersonal art form. When we offload teaching to an algorithm, we aren’t personalising education; we are isolating the student.
AI is a brilliant tool for narrow tasks. It can challenge a thesis or help debug code at 11:00 PM, but it cannot mentor a child. It can support a child with a specific learning need probably because it can provide the personalised assistance that a teacher cannot offer in a crowded Indian government-aided classroom, or because the teacher is not available when the child needs her. But AI cannot hear the subtle quiver in a student’s voice that signals a loss of confidence. It cannot facilitate the messy, human friction of a classroom debate where true democracy is born.
This is where India’s National Education Policy (NEP) 2020 gets it right; it explicitly anchors this tech-heavy future in teacher empowerment. The NEP recognizes that technology must reduce administrative drudgery so that teachers can focus on high-value human interaction. It mandates a shift from a teacher being a deliverer of facts to a mentor who navigates the chaos of the digital age alongside the student.
Our goal must be to cultivate problem solvers. AI is not going to solve those problems. AI-educated children much less so. In fact, an MIT Media Lab Study in 2025 showed that 78% of MIT students became so reliant on AI they couldn’t remember their own writing after 4 months. AI can give a student the answer, but only a teacher sitting by their side can teach them why that answer matters, how it impacts society, how to verify if it’s even true, and most importantly, how to work with others to apply that answer to solve society’s problems.
Recently concluded and globally promoted AI Impact Summit in India also disproportionately extolled the virtue of AI without investigating the risks AI brings in cognitive decline. On the other hand, the Davos panel made it clear: in an age of infinite artificial intelligence, human intelligence is our most scarce and precious resource. We need teachers more than ever to act as the cognitive and emotional anchor.
The mission, now, is to use digital and AI in the classroom to invest in the only ‘technology’ that has ever actually worked: the transformative power of one human being guiding another to solve our shared problems.
