AI and the Future of Skilling | India AI Impact Summit 2026

Contents

Executive Summary

This panel discussion addresses how AI will transform higher education and workforce development in India, emphasizing that disruptions like AI create new industries and competencies rather than eliminating them entirely. The panelists argue that educational institutions must shift from factory-model, degree-focused systems toward competency-based, adaptive learning powered by AI infrastructure, while maintaining human-centered skills like emotional intelligence and critical thinking that machines cannot replicate.

Key Takeaways

AI is not about replacing humans—it's about upgrading institutions: The real challenge is transforming educational systems from factory models (standardized degrees, passive learning) to adaptive, competency-based models that develop human capacities AI cannot replicate (judgment, empathy, creativity).
Competency-based, continuously updated curricula are non-negotiable: Static 4-year degree programs cannot survive Industry 4.0. Institutions must map syllabi to industry competencies and refresh them every semester, especially in creative and technical fields.
Build trust infrastructure (DPI), not just technology: India's advantage lies not in winning the AI model race but in creating shared, open digital infrastructure that lets millions of educators, entrepreneurs, and learners build solutions. Trust, verification, and discovery mechanisms (the "roads") matter more than the specific "cars" (AI models).
Emotional, behavioral, and foundational human skills are the new scarcity: Teach grammar (foundational creative arts, critical thinking, ethics) early; let AI handle rapid tool updates. Emotional intelligence, storytelling ability, and moral reasoning are the careers AI won't disrupt.
Decentralization and agency matter more than access: The goal isn't getting a billion people access to one Nobel Prize–quality AI. It's enabling 1.5 billion Indians to become AI agents in their own right—creating their own micro-AIs, solving local problems, and building livelihoods tied to their identity and autonomy.

Key Topics Covered

Historical context of technological disruption — Industrial Revolution, computers, internet, and AI as cyclical forces that create more jobs than they eliminate
Quality at scale — How to deliver MIT-quality education to billions of learners through modular, AI-powered platforms
Competency-based learning — Aligning syllabi with industry-demanded competencies rather than static degrees
Creative industries and storytelling — AI's role in AVGC (Animation, Visual Effects, Gaming, Comics) and extended reality; emotional intelligence as irreplaceable
Digital Public Infrastructure (DPI) — Creating shared, trustworthy infrastructure (like Aadhaar, UPI) that enables education innovation at nonlinear scale
Decentralized AI agents — Moving from centralized AI models toward distributed AI agents that empower individuals and communities
Intergenerational learning — Pairing younger people who know how to use tools with older people who know what to do
Foundational pathways for creative arts — Creating primary/secondary curriculum for design, storytelling, film, and performing arts
The skills-motivation gap — Youngsters can use tools but don't know what to do; education must bridge knowing how with knowing why
Nonlinear solutions for nonlinear problems — India's gap between rich and poor is growing nonlinearly; solutions must scale nonlinearly

Key Points & Insights

Educational lag behind industry: Education systems remain at "Education 2.0" standards while industry has advanced to "Industry 4.0," creating a dangerous mismatch that AI intensifies rather than solves.
AI as a forcing function for institutional reflection: Rather than being about AI itself, the disruption requires educational institutions to clarify what they truly value — active learning, practical engagement, interdisciplinary problem-solving — and redesign pedagogy around these invariants.
Scale is a vector, not a scalar: MIT's "open courseware" model of 1999 (distributing one thing to millions) has evolved into "universal AI" — understanding that different learners need different pathways, motivations, and timings. Scale now means "many things for many different kinds of people."
Emotional and behavioral intelligence are non-fungible: AI excels at content creation, editing, and execution, but storytelling, filmmaking, and teaching require emotional evolution and empathy. The role of "human in the loop" (like editor Raj Kumar Hirani) remains essential, though this may eventually shift.
Competency frameworks must be continuously updated: Because industry itself struggles to predict what skills will be needed in 2-3 years, competency-based curricula must become living, adaptive documents tied to real-time industry feedback — not static 4-year degree programs.
Foundational skills vs. cutting-edge skills split: Primary/secondary education should teach foundational grammar (storytelling, design, creative arts, STEM basics), while higher education and industry partnerships handle rapid tool updates and specialization.
Digital Public Infrastructure unlocks nonlinear solutions: DPI models (like Aadhaar for identity, UPI for payments) create shared foundations that allow thousands of innovators to build educational solutions on top, rather than centralized institutions monopolizing delivery.
The "bazar era" of AI is emerging: Moving away from centralized AI factories (expensive, few models, many users downstream) toward a "garage era" (anyone builds AI) and eventually a "bazar era" (any individual creates and trains micro-AI models as part of livelihood and identity).
The dystopia risk: intelligence slavery: If only a handful of mega-models exist (one math teacher AI, one doctor AI), even employed people become soulless functionaries delivering pre-built intelligence rather than thinking agents — a risk that decentralized, local AI agents mitigate.
Passion and motivation are now career prerequisites: Unlike previous generations, entering a field you're not passionate about may lead to a career that doesn't exist by graduation. Education must enable discovery of passion alongside skill-building.

Notable Quotes or Statements

Dr. Vijay Kumar (MIT): "Scale is not a scalar. It's a vector. You know it's got magnitude and direction." — Highlighting that modern educational scale means serving diverse learners with different needs, not broadcasting one-size-fits-all content.
Ashish Kulkarni (IICT): "Every semester will have to be a different semester because of AI... the foundational skills are the ones that you can actually look at going early." — Emphasizing the need for continuous curriculum redesign.
Dr. Manish Kumar: "Industry is at 4.0 and education is still at 2.0... the desire for predictability and stability within the educational system [conflicts with] the forward move of industry which you cannot stop." — Diagnosing the core institutional lag.
Shankar Maruada (XStep/Aadhaar): "When problems are growing nonlinearly, our solutions cannot grow linearly... We need our solutions to also grow nonlinearly." — Core principle behind DPI-based skilling.
Prof. Romesh Razar (MIT Media Lab): "We're going to move into this I would call intelligence slavery where even if people don't lose their jobs they'll be soulless heartless jobs... The only country that has a bazar model already figured out... is [India]." — Describing the centralization risk and India's DPI advantage.
Shankar Maruada: "If I went somewhere and got a skilling course on editing, how do I know it's genuine? So you need to create reduce the cost of trust in the digital world." — Highlighting why trust/verification infrastructure is critical to DPI-based education.

Speakers & Organizations Mentioned

Primary Panelists

Dr. Vijay Kumar — MIT, Council on Education Technology; leading Open Courseware and Universal AI initiatives
Ashish Kulkarni — Founder-Director, Indian Institute of Creative Technologies (IICT); AVGC/XR pioneer
Dr. Manish Kumar — Former Managing Director, NSDC; administrative services background; institutional economics expert
Shankar Maruada — Co-founder, XStep Foundation; architect of Aadhaar digital identity system
Prof. Romesh Razar — MIT Media Lab; leading Internet of AI Agents research

Key Institutions & Initiatives

MIT — Open Courseware (1999); Universal AI platform
Indian Institute of Creative Technologies (IICT) — New institution focused on AVGC, storytelling, design
NSDC — National Skill Development Council
XStep Foundation — Digital public infrastructure and skilling
Aadhaar — India's biometric identity system (1.5+ billion users)
UPI — Unified Payments Interface (digital payment infrastructure)
Project Nanda — Network AI Agents in Decentralized Architecture (MIT Media Lab research)
Lauren Musicacademy.ai — AI-powered music learning platform (co-created by Lauren and Aubrey, mentioned as India-based success)
MIT Media Lab — Decentralized technologies and AI agent research
KPMG — Government and public sector advisory (co-hosting/sponsoring)

Government & Policy References

National Education Policy 2020 (NEP) — Mentioned as forward-looking but still adapting to change
Indian Budget references — IICT allocation; Creative Arts/Performing Arts curriculum integration
Tinkering Labs — 10,000 labs expanding to 50,000; creative STEM spaces

Technical Concepts & Resources

Platforms & Infrastructure

Universal AI — MIT's modular, interactive foundational AI courses with adaptive pathways; targeting 1 billion learners
Open Courseware (MIT, 1999) — Free publishing of course content and resources; foundation for modern MOOC movement
Aadhaar — Biometric digital identity infrastructure enabling nonlinear service delivery
UPI — Payment infrastructure enabling gig economy and financial inclusion
Digital Locker — Document/credential storage and verification system
DigiYatra — Airport digital identity verification (mentioned as DPI application)
Project Nanda / Internet of AI Agents — Decentralized, agent-based architecture allowing individuals to create and deploy micro-AI models

Learning & Pedagogical Concepts

Competency-based learning — Mapping curriculum to industry-demanded competencies rather than static degrees; continuous updating
Active learning — Hands-on, problem-based learning (MIT's "minds on, hands on")
Intergenerational learning — Informal mentorship between "people who know how" (young technologists) and "people who know what" (experienced practitioners)
Formative testing with AI — Using AI to continuously assess learner progress and adapt pathways
Differentiated learning pathways — AI-driven personalization using learning science (learning curves, forgetting curves)
Interdisciplinary/cross-disciplinary learning — Addressing complex problems (climate, health, AI ethics) that require boundary-transcending teams

Foundational Curriculum Areas

AVGC (Animation, Visual Effects, Gaming, Comics) + XR (Extended Reality) — Including passive (film/web series), interactive (gaming), and immersive (XR) storytelling
Creative Arts & Performing Arts — Now being integrated into NEP 2020 grades 6-12; foundational for design/storytelling pathways
Design, Storytelling, Film, Sports — Alternative career pathways alongside STEM
Emotional Intelligence & Behavioral Intelligence — Non-replicated human competencies; critical in media/entertainment context

Trust & Verification Infrastructure

Verifiable Credentials — Digital proof of genuine skills/qualifications (reducing cost of trust in DPI)
Orange Economy — Content creator/creator economy enabled by platforms; discussed as emerging livelihood model
Gig Economy Infrastructure — DPI enabling on-demand work and income (e.g., delivery, content creation)

AI/ML Concepts Mentioned

General Purpose Technology (GPT) — AI positioned alongside steam, electricity, internet as civilization-altering technologies
Micro-AIs / AI Agents — Individual-created, small-scale AI models serving local/personal purposes (vs. mega-models)
Machine Teaching — AI tutors and pedagogical agents providing personalized guidance
Decentralized AI — Multiple independent AI agents coordinating (vs. centralized mega-models)

Tools & Platforms Referenced

Replit — AI-assisted coding platform (example of rapid tool improvement in AI era; was weak 1 year ago, now advanced)
PowerPoint + AI — Example of how tools remain personal/differentiated even when commercialized centrally
Copilot / AI Coding Assistants — Enabling faster skill acquisition for programming

Important Contextual Notes

Session format: 55-minute panel + intended audience Q&A (though time ran out for audience participation)
Summit context: Opening session of India AI Impact Summit 2026 (5-day event)
Language & accessibility: Discussion emphasizes need for Indic language AI support to overcome language barriers to skilling
Demographic focus: India's 1.5+ billion population; focus on rural, low-income, and underserved populations; mention of villages, girls' education, gender equity
Policy context: NEP 2020 adoption; Government of India's 15,000-school rollout plan for creative arts curriculum; 50,000 Tinkering Labs expansion
Philosophical underpinning: "Build roads, not cars" — create DPI infrastructure and let market/innovators build solutions on top
Time reference: Discussion positions 2026-2047 as critical window for preparing youth for Viksit Bharat ("Developed India") goal