We are living in an exciting time when just one technological breakthrough can transform the world as we know it. The spectacular rate of innovation means that our lives, once again, will look very different in just a generation or two. I’m talking about AI, of course.
Most enterprise companies today are wondering whether AI is worth the hype, especially since the late 2022 release of ChatGPT opened the floodgates of generative AI. As a tech company CEO who specializes in helping companies ramp up growth and productivity, I agree with the words of my company’s Chief Technology Officer, Elise Carmichael: “We’ve all heard that ‘AI is eating the world.’ We don’t need to fear it, however. Instead, why not grab a spoon and dig in?”
I’ll even add, “Let’s grab a seat at the table now.” Why? Because AI, including ML and well-trained generative AI models, can take complex enterprises to their next level of success. By “grab a seat at the table,” I mean going beyond the tactical and simple use cases and addressing AI at the business strategy level. It is my view that the companies that seize the opportunities of AI now will be the ones that ultimately thrive.
I believe enterprises must filter their AI strategy through three crucial lenses in order for this competitive edge to come to fruition down the road. These lenses are:
1. RELEVANT AI USE CASES
Almost all successful products or solutions are invented to solve a known problem. If we think of AI as another breakthrough product, we can vet it with many of the same questions a strong product leader should ask: What is the specific problem (or problems) it is trying to solve? Does it deliver measurable outcomes for targeted customers? Can it scale in the market?
In my field of IT, generative AI models could enable autonomous IT support—where a well-trained AI model can detect and resolve level-one service tickets without human intervention, freeing up technicians for more strategic work. All companies working in the digital employee experience (DEX) space are racing to capture this very important use case.
Another important use case is reducing mean time to resolution (MTTR) through anomaly detection, which is a compelling use case for machine learning models. These models are known to enable predictive analytics that can minimize costly downtime and unnecessary hardware refreshes by accurately anticipating when a device is likely to fail.
2. DATA TO FEED THE AI
One aspect that almost all executives I speak with agree on is that any AI model is only as good as the data feeding it. As most AI experts will attest, building a baseline generative AI model, as complex as that legwork is, is less challenging than training and finetuning the model. Why? Because generative AI models need robust data sets to tone down their notorious hallucinations or to make them serve up relevant answers that are explainable.
The challenge for many enterprises is that they often do not have access to years of valuable data. The lack of good data to train the models is an undeniable challenge. Indeed, data will make or break a company’s ability to apply AI models to concrete use cases.
A robust data strategy is, therefore, inextricable from a successful AI strategy. An AI model without good data (both quality and quantity) is just another shiny object without much substance. Not only does the depth and breadth of data matter, but so does data history. Combined, this trio can deliver the data fidelity needed for AI models to output reliable results. That brings me to my next point: trust.
3. TRUST IN AI
I suspect that trust in AI models will remain a legitimate concern for many years to come. While better data can lead to better results, it will take many organizations quite some time to become comfortable with AI in general.
Soon, generative AI models that are fed the right data will be as good as—if not better than—most systems engineers at resolving problems. Still, keeping the “human in the loop” is imperative for building models that are trustworthy enough to co-pilot use cases for IT and other enterprise functions.
WELCOMING AI INNOVATION
“Humanity never thrives when it fears innovation,” notes a recent HBR article. As responsible AI use cases are being developed fast and furiously, let’s pull up a seat and celebrate the transformative possibilities of inevitable AI.
As CEO of Lakeside Software, David Keil has a distinguished record of leading enterprise IT organizations through exceptional growth.