After spending five months as a digital transformation consultant in the Indian AI space, I have had the opportunity to observe this sector from both the ground level and the strategy deck. It’s been long enough to move past first impressions, and short enough to still feel the pulse of experimentation. What I’ve seen is a landscape full of potential, but also full of paradoxes.
There is a curious duality here, a bimodal distribution of excitement, as Ani Kelkar aptly puts it in his latest think piece. On one end, there’s a group that sees AI as a revolutionary force, diving deep into its architecture, discussing transformer models, fine-tuning, and prompt engineering. On the other, there are those who treat AI tools like glorified search engines – useful, but not transformative.
This isn’t just a philosophical divide; it’s a practical one. And frankly, it’s not unique to AI. This pattern is something I have seen across most emergent technologies (Robotics, 3D Printing, Blockchain) where early adopters and skeptics coexist in a strange equilibrium.
AI is Everywhere… and Nowhere
AI is now a staple in SaaS product pitches, keynote speeches, and panel discussions. It’s peppered into conversations like seasoning; always present, but rarely the main course. Despite all the talk, I have yet to see AI meaningfully embedded in the day-to-day operations of most companies.
Yes, we’re (India) building semiconductor fabs. Yes, we’re debating which GPUs to import. But the real question is: what are we building them for?
The Missing Link: Use Cases
The biggest gap isn’t infrastructure, but imagination. We’re (India) not short on compute power or talent. What we’re short on are use cases grounded in business reality. Many of the AI case studies I hear about sound impressive, but I often find myself questioning their authenticity. Not because I think they’re fabricated, but because they feel disconnected from the ground truth.
Most of the developers I speak with still rely on good old Python, C#, and Java. When they do try AI-assisted coding, what some call “vibe coding”, it often creates more work than it saves.
Sector Watch: Where’s the Adoption?
While the tech sector has seen exponential adoption of AI-ML powered applications, other industries are lagging. In BFSI, AI is mostly limited to fraud detection and chatbots. In healthcare, it’s used for diagnostics and imaging, but adoption is uneven and often pilot-level.
And then there are the two Indian biggies: Agriculture and Manufacturing.
These sectors contribute over 35% of India’s GDP and employ more than 60% of the workforce. Yet, AI adoption here is minimal. Why? Because AI isn’t perceived as affecting daily bread-and-butter operations. The motivation to adopt is low, and the perceived risk is high.
Sure, you’ll see branding, marketing campaigns, and promotional material claiming AI integration. But that’s a far cry from reality. Until sectors like agriculture and manufacturing integrate this technology in earnest, we won’t see meaningful growth. These sectors produce actual, tangible, measurable, and impactful products, making them drivers of real economic progress.
So, What’s Really Happening?
It’s a muddy scene. There’s genuine interest, but also confusion and uncertainty. People want to use AI, but they just don’t know how or where to start. And that’s okay. Every technological shift begins with uncertainty.
Here’s a practical suggestion: start small, start internally.
Run AI training sessions for routine departments such as accounting, procurement, and admin. Get people talking. Let them explore how AI could help with their daily tasks. These conversations will naturally lead to real, grounded use cases. And once those emerge, the tech can be tailored to fit.
AI doesn’t need to be a moonshot. It can start with a spreadsheet, a workflow, or a dashboard. The key is to move from discussion to deployment; from hype to habit.
Feel free to share your thoughts in the comments below!
