Nilesh Khedkar.
← All writing

Why AI Adoption Is Still Low in enterprises

I still remember those long bus rides through northern Uttar Pradesh — the Babrala ( a village in northern India) to Agra route on UP Roadways. Dusty roads, blaring horns, chai…

Nilesh Khedkar · 4 Dec 2025 · 3 min read · AI, Generative AI Use Cases, Applied AI

A Story from a UP Roadways Bus

I still remember those long bus rides through northern Uttar Pradesh — the Babrala ( a village in northern India) to Agra route on UP Roadways. Dusty roads, blaring horns, chai stops, and the occasional surprise salesmen who turned the bus aisle into a mini marketplace.

Among them was one unforgettable character:

the lemon-squeezer salesman.

The Magician of the Bus Aisle

He had a routine.
He would hop on at one stop, walk confidently to the center of the bus, and start his show.

With dramatic flair, he’d slice a lemon, position it in the device, and press.

One smooth motion. Perfect juice. No mess.

Passengers watched wide-eyed.
People nudged each other and whispered, “Wah! Kitna aasaan hai!”
(“Wow! That’s so easy!”)

And within a few minutes, half the bus — including me — felt like life would be incomplete without this magical tool.

I bought one too.

The salesman would get off at the next stop, money in hand, leaving us with dreams of effortless lemon juice forever.

The Return to Reality

But as soon as I reached home, I learned something important:

  • I hardly ever made lemon juice.
  • The “effortless” squeeze he showed required actual effort.
  • And without the perfect bus-demo conditions, it didn’t feel magical at all.

The tool wasn’t bad.
It just didn’t fit into my actual lifestyle.

The problem was not the product — it was the assumption that I needed it just because the demo looked impressive.

And This… is Today’s AI Story

Fast forward to today — and companies everywhere are repeating the same pattern.

AI demos wowed the world.
They were smooth, polished, magical — just like that lemon-squeezer sales pitch.

Organizations rushed to buy licenses and platforms because they didn’t want to “miss out.”
Then came the hackathons, the workshops, the internal webinars… all trying to answer one big question:

“Now that we have AI, what do we do with it?”

It was the same mistake:
We bought the tool before knowing whether we actually needed it.

Employees didn’t understand it.
Leaders couldn’t fully explain it.
And daily workflows didn’t match the shiny demos.

So adoption slowed.
Not because AI failed — 
but because the fit wasn’t there.

The Real Lesson From the Bus Ride

The lemon squeezer didn’t change my life because it wasn’t connected to my real habits and needs.

AI won’t change an organization unless it’s connected to real processes and real pain points.

The demo is just a performance.
The reality is in the daily work.

How AI Truly Gets Adopted

For AI to succeed, we need to reverse the order:

1. Start With the Problem

Understand the workflow, the bottleneck, the frustration.
Only then decide if AI can help.

2. Make Users the Heroes

People don’t reject AI — they reject confusion.
Explain it. Show it simply. Connect it to their work.

3. Introduce AI Step-by-Step

Not as a magic gadget,
but as a thoughtful change in how work gets done.

Magic happens in presentations.
Real results happen only in daily routines.

AI Doesn’t Fail. Our Approach Does.

Just like my lemon-squeezer from that UP bus ride, AI is dazzling in the demo.

But true adoption happens only when the tool fits the people, the process, and the problem.

If we want AI to succeed, we shouldn’t sell dreams.
We should build understanding.

And maybe that’s the real lesson the UP Roadways bus taught me — long before AI ever entered the conversation.

Originally published on Medium.

Say hello

Always happy to swap notes on AI in industry, maintenance and reliability.