"No, this is not an AI project."
"No, we don’t need an LLM to solve this."
"No. No. No."
If AI seems like the answer to every question, then your AI specialists might not be what they seem to be.
True AI expertise is not about building models. It’s about deciding when not to.
Simple answer saves time, money, and resources
📦 Example 1 — Counting Boxes on a Pallet
A request came in:
“Can we use AI to count the number of boxes on a pallet?”
It sounded futuristic — but completely missed the point.
We took a step back and had a chat with the people on the ground floor. Their current method? Simple and elegant:
- Weigh the full pallet.
- Weigh one box (the weight is written on it).
- Divide one by the other.
✅ Result: accurate count, no cameras, no cloud infrastructure, no computer vision, no problem.
Sometimes the most effective AI project is no AI project at all.
Simple weighing method beats complex computer vision for counting boxes
☕ Example 2 — The Coffee Machine
Every coffee machine checks if a cup is placed under the nozzle before pouring. You could use computer vision to detect the cup and its position. Or... You could use a simple sensor.
If sensor == TRUE: pour. If sensor == FALSE: don’t pour.
✅ Result: simple, robust, low-cost.
Simple sensor reliably detects cup presence, no AI needed
💡 The Deeper Lesson
Saying “No” is not a lack of innovation. It’s a sign of experience and maturity.
Every unnecessary AI project consumes:
- Engineering effort
- Budget
- Cloud resources
- Focus from what truly matters
The ability to say “No” is what differentiates an AI enthusiast from an AI expert.
🧭 Closing Thought
Every time you say “No” to an impractical AI project, you say “Yes” to scaling the right ones faster and smarter.
True AI leadership is not in chasing every use case —
It’s in recognizing which ones aren’t worth chasing.
Tags: #AI #MachineLearning #DigitalTransformation #Leadership #Experience #ResponsibleAI