My Tryst with Statistics and AI

An AI generated image

With all the buzz around AI these days, I figured it was time to jot down my thoughts on the journey—from hating statistics to embracing AI.

 The Early 1980s: Hating the "Magic"

Back in my engineering college days, I despised statistics. It felt like some magician's trick: "There are 5 black and 8 white balls in a bag. Pick one at random—what's the chance it's white?" Pulling rabbits, pigeons, or balls from bags? No thanks. I deliberately dodged it.

Mid-1980s: Thrown into the Deep End

Then I joined MICO (now Bosch) as a trainee engineer and rose to become an Inspection Foreman. Suddenly, I was swimming in that "magic" sea. No desktops yet those day—just mainframes like Data General or IBM (too big for our small tasks) or we crunched data manually: plotted histograms, frequency charts, used log-normal plots to calculate averages (X Bar) and standard deviations (s or sigma). Probability helped predict defects and decide whether to accept or reject production lots.

That's when statistics clicked. I dove in headfirst, enrolling in a Statistical Quality Control diploma at the Indian Statistical Institute. Our faculty covered how to distribute t or Poisson (t-distributions, Poisson distribution), F-tests, chi-square, orthogonal arrays, design of experiments, and hypothesis testing. Beyond the jargon, I fell in love with *prediction*. I used it to optimize processes and stabilize manufacturing.

Late 1990s: A Hiatus and a Comeback

I took a break for finance, accounting, and control roles. Returning to materials management (purchasing, inventory, logistics) reignited my interest—especially with Six Sigma all the rage, thanks to GE. Since then, I've dipped in and out, sniffing around stats for analysis and decisions.

2020: COVID and a Fresh Start

That grim COVID year locked me down and freed up time. I'd heard AI and ML chatter but ignored it amid business demands. Homebound, I thought: Why not learn AI?

Serendipity struck—my school friend, Dr. Gopalakrishna Sharma, was teaching it at IISc's Continuing Education Program. I begged him to be my guru; he agreed. When he asked my stats level, I admitted: 35 years rusty. We started from scratch—data basics, histograms, central tendencies, dispersion—and built to advanced topics. That's how I grasped AI and ML.

What Is AI? (In Layman's Terms)

Artificial Intelligence (AI) lets computers tackle tasks by mimicking smart humans, like learning, reasoning, or spotting patterns.

Core Idea

AI mimics human thinking by crunching massive data, finding trends, and deciding without rigid rules. Unlike old software, it learns from experience—like a kid through trial and error.

Natural vs. Artificial Intelligence

Human (natural) intelligence draws from biology, emotions, and senses for creative adaptation. AI uses algorithms and data for blazing speed and scale—billions of calculations in seconds.

Everyday Example

Your phone's Siri: It hears you (perception), gets your intent (reasoning), and replies (decision)—trained on millions of chats. Contrast that with chatting fluidly with a friend, weaving in context and emotion.

AI: Friend or Foe?

Debates rage on ethics and morality. My view? AI's a tool—like a knife. Wield it for good or harm. Embrace it, adapt, and level up. Creativity? AI remixes what's out there; it can't yet be a Picasso, Shakespeare, Kalidasa, or Ravi Varma.

 How I Use AI

I stick to tools like ChatGPT, Copilot, Canva and Perplexity for research (corporate, legal, comparisons), editing writing, generating images, and more. No Python coding for me—I'm happy boosting what I do by using AI tools.

AI and Me

All said and done, I'm Nagamangala Srinivasan Mohan—or N.S. Mohan. I like being "NS," but I won't be "Naturally Stupid" to AI.

What's your take on AI? Are you using it? If yes, how? Feel free to comment.