The AI skills gap in India is no longer about access to courses. It’s about depth. Reports say about 20 lakh Indians have been upskilled for AI, but only around 3 lakh have advanced skills. That points to a pipeline that is wide at the top and thin where it counts. Few people have thought harder about this gap than Jaspreet Bindra. He is founder of AI&Beyond and Tech Whisperer Ltd, and author of Winning with AI: Your Guide to AI Literacy. His argument is simple and uncomfortable: India has been counting trainees when it should be building literacy.
Here is what Bindra believes, and what it means for jobs, employers and policymakers.
Who Is Jaspreet Bindra?
Bindra brings a practitioner’s lens, not just a commentator’s. He has served as group chief digital officer at the Mahindra Group, as a regional director at Microsoft India, and as a general manager in the Tata Group. His second book, Winning with AI, published in 2025 by Juggernaut, argues that AI literacy is the essential skill of this era. He also holds a master’s in AI ethics and society from Cambridge. carnegieendowmentcarnegieendowment
Why the AI Skills Gap in India Is a Literacy Problem, Not a Training Problem
Bindra’s core point is that training and literacy are different things. Training teaches you how to use a specific tool, while literacy gives you the grammar to work with any tool, in any context. His analogy is easy to remember. Once you are literate in English, you don’t need training for every novel you read. carnegieendowmentcarnegieendowment
That difference explains much of the gap between 20 lakh and 3 lakh. A short course that teaches people where to click inside one AI tool produces “trained” workers. It does not produce people who can adapt as tools change every few months. Bindra sums up the goal in his first “law” of AI literacy: not everyone needs to be an expert, but everyone needs to be AI literate. carnegieendowment
He also widens the definition of AI talent. In his view, AI talent won’t just mean ML engineers and data scientists. It will mean anyone who uses AI to do their own job better. His example is a recruiter who automates their own hiring workflow with no-code AI tools instead of waiting on IT. carnegieendowment
The READS Ladder: How Bindra Defines “Advanced”
If basic upskilling is the first rung, what does “advanced” look like? Bindra and his AI&Beyond co-founder Anuj built a five-step ladder that runs from reads to writes to adds to thinks to does. carnegieendowment
- Reads: absorbing information faster, for example by summarising with AI
- Writes: creating content across text, video and other formats
- Adds: analysing data with AI
- Thinks: using AI as a second brain to brainstorm new ideas carnegieendowment
- Does: using agentic AI to make AI actually carry out tasks carnegieendowment
Most of India’s 20 lakh upskilled workers probably sit on the first two rungs. The 3 lakh with advanced skills are the ones climbing toward “thinks” and “does.” Bindra also says the ladder is getting taller. He plans to add steps such as making, drawing and orchestrating, which agentic AI now demands. carnegieendowment
What the AI Skills Gap Means for Jobs
Bindra doesn’t sugar-coat disruption. He accepts that AI will destroy certain jobs and create new ones, and that the rupture will be real. But he thinks “job destruction” is the wrong frame. carnegieendowment
He points to the history of Indian IT. The sector was itself born from the last great technology disruption, when fear that computers would wipe out clerical work gave way to more fulfilling, higher-paying jobs. Stenographers disappeared, and an entire industry appeared in their place. carnegieendowment
His conclusion is that India should protect workers, not jobs. Trying to protect jobs is futile, he argues, but workers who reskill through AI literacy can move into newer roles. This is why the 3 lakh number matters. The workers who climb the ladder move into the jobs AI creates. Those who stop at basic training risk being stuck in the jobs it removes. carnegieendowment
He also names the sectors under the most pressure. Coding and software engineering show the highest AI usage. Customer service is catching up. BFSI is eager but held back by regulation. Consulting faces an existential threat, or at least a disrupted business model. carnegieendowment
The Real Barriers: Fear, Cost and Awareness
Why don’t trained workers progress further? Bindra’s answer is mostly psychological. Corporate professionals feel overwhelmed by the hype. Many stick to free tools, run into hallucinations, and won’t pay for better versions. He calls fear the biggest barrier, driven by stories about deepfakes, data leaks and environmental harm. carnegieendowmentcarnegieendowment
The adoption gap inside companies is just as stark. Even three years after ChatGPT, he finds enterprise adoption abysmally low. In his words, organisations have brought the horse to the water but can’t make it drink. He links this to a wider trend. He cites the MIT NANDA finding that 95% of generative AI startups don’t scale, partly because people weren’t literate enough to adopt them. carnegieendowmentcarnegieendowment
Beyond the corporate world, the gap is bigger still. He estimates that over 80% of Indians may not even know AI tools exist. carnegieendowment
Bindra’s Prescription: Treat AI Literacy as a National Mission
Bindra supports what is already happening. That includes AI becoming a separate CBSE subject and Budget 2026’s proposal for 15,000 school AI labs and 10,000 IIT tech fellowships. His own firm contributed too. AI&Beyond made the government’s “YUVA AI for All” course, which he says close to half a million people have registered for. carnegieendowmentcarnegieendowment
But he argues India needs a bigger shift in approach. He wants AI treated as a full national mission, like the Green Revolution or digital public infrastructure, rather than a scattered set of schemes. He calls this vision “JanAI”: AI for the janata, delivered as a digital public good through the apps and devices people already use. carnegieendowmentcarnegieendowment
The numbers suggest he has a point. Even earlier government programmes showed the same drop-off. In 2024, MeitY reported that over 18.56 lakh candidates had signed up on its FutureSkills PRIME portal, but just over 3.37 lakh had completed a course. Enrolment has never been India’s problem. Depth has. ianslive
The Bottom Line
The AI skills gap in India won’t close by adding more names to training dashboards. Bindra’s view, in short:
- Measure literacy, not attendance.
- Help people climb from reading with AI to acting with it.
- Protect workers rather than jobs.
- Give AI the national-mission scale that made India’s digital public infrastructure a global model.
For professionals, the takeaway is personal. A certificate doesn’t mean you’re ready. Fluency across tools does.