“You have been doing AI for quite some time now, before it was the trend of 2023, how has AI developed to be at the level it is now?”

I was asked this question during The New Warehouse podcast episode, which will debut in mid-March.

“AI” is quite the buzzword these days and has generated significant popularity among the masses, probably coming in second or third place behind Taylor Swift or the Kardashians—all thanks to the introduction of OpenAI, particularly ChatGPT, to the world over a year ago. Now, “the horse has left the barn,” and we cannot escape AI’s wrath, good or bad, depending on who you ask.

Go on LinkedIn, and there’s a proliferation of profiles with AI-expert, AI-guy/gal, or AI-consultant, all within the last year. Turn on the TV, and you’ll see mainstream media pontificating about it. I compare this to Tom Brady declaring himself the GOAT after winning his first Super Bowl. Sit down and try to have an intellectual conversation with these newly-minted, self-declared GOATs of AI on predictive, prescriptive analytics, the types of algorithms they use for their analysis, or the business problems they solve, and they’ll typically chirp back: ChatGPT, AI is taking my job, or you’re going to ruin humanity.

This year’s NRF had grown men parading in t-shirts emblazoned with “I AM AI” (what does that even mean?)across their chests or, since 2023, companies slapping the “AI sticker” on their marketing collateral. Separating the wheat from the chaff from those AI software companies who peddle a plugin or layer on top of OpenAI (glomming on someone else’s tech, masquerading as theirs without licensing rights) or selling another ChatGPT instruction manual can be a daunting task.

How do business leaders of multi-million dollar companies who want to improve their top and bottom lines with the help of AI consulting and software deal with the “AI hype” as inflation, high interest rates, and labor shortages continue to hurt the economic landscape?

It’s All About Good, Clean, First-Party Data

A solid definition of AI (though there is no universally accepted meaning) is getting a computer to think and act like a human. Thanks to Alan Turing, AI has been around since the 1950s and has been part of our everyday lives for the last decade with Amazon and Netflix products, movie recommendations, and our beloved Alexa and Waze, which mostly give us accurate driving directions. However, it wasn’t until the launch of ChatGPT that almost everyone became aware of the concept of AI. It doesn’t matter what AI software solution you use; AI is about analyzing past patterns to make predictions.

With advancements in technology, digital transformation, cloud infrastructure, and the advent of the smartphone, AI software powered by machine learning algorithms has access to vast amounts of data. However, data isn’t just data. You’ll need solid, clean data to make accurate predictions that help solve your business challenges. Without this, your AI isn’t as powerful, leading us to the ChatGPT problem, which relies on open-source LLM models that can give its users false hopes based on the output they provide despite paying $999 for the latest ChatGPT “How to Manual.” Even if you feed it your data, its capacity can be limited.

Solving business problems can be complex. There’s no easy button. It takes well-thought-out strategies, cleansing data protocols, and robust analytics to make predictions based on studying your first-party data that drive real business value.

Typical pain points that AI can help solve include:

  • How to reduce stock dead stock percentage by 5%
  • How to improve average order value by 8%
  • How to predict the potential profitability of sleeping customers to reduce churn by 4.5%

Notice that the numbers aren’t spectacular, negating the AI hype factor you may see on LinkedIn, tradeshows, and mainstream media. This example is significant because, like in football, AI is a game of inches, getting smarter every time to improve a business’s profitability over the long haul.

ChatGPT Et Al. Isn’t All Bad

What do you call the student who finishes last in med school? Doctor.

What do you call the person drafted last in the NFL? Mr. Irrelevant.

What do you call the lowest-ranked person in the graduating class at West Point? The goat.

All examples are significant accomplishments – well ahead of almost everyone.

I’m in my early fifties, and when I was in high school in the late eighties and failed to do the required reading for my Youth In Literature class, I would resort to buying Cliff’s Notes at the local book store (there was no internet or Amazon back then) to get a handle on the missed reading assignment. It was better than nothing; I would get a C, failing to live up to my full potential of getting an A, if I only had done the required reading.

It’s great that AI is now mainstream – using it will put your business ahead of most. However, using only ChatGPT or other AI tools that fail to address your business problems is as good as getting a C – better than nothing, but not as good as getting an A.

Don’t fall prey to charlatans like the main character in the child’s storyThe Emperor’s New Clothes. Integrating AI into your business processes is challenging and will require time and resources. Do the work. It won’t be an overnight success. However, if you approach it correctly, with the right mindset and tools, you can build your company’s economic moat with the help of AI.

Now, for those in the back – when I say “A,” you say “I” – A…I…A…I