The What, Why, Where and How of the AI Conversation
“We use AI.” “It’s AI-based.” “AI-powered solution.”
If CEOs, CIOs, CFOs and all their managers had a nickel for every time these statements or some variation was declared, we could solve our national debt. AI misstatements and overt declarations are pervasive in today’s landscape with the term being used broadly to mean many things. A technologist may have one understanding, a product manager another and a salesperson might have her own interpretation. Every day, I see and hear misunderstandings due to our broad, and mostly disparate, understanding of the term.
Here are some clear and focused guidelines for discussion to help provide a response to the inevitable phrase, “We use AI.” In this article, I’ll use the “What, Why, Where, When, How” sequence to provide an outline to help facilitate conversations with your AI software vendors and consultants.
What is AI?
The concept of Artificial Intelligence (AI) was brought about by scientists and mathematicians in the 1950s who explored the mathematical possibilities of machines making decisions with human-like intelligence. At that time, computers could not store commands, they just executed instruction sets. But the dream and concept of AI was alive and kicking: it was the ability of machines to perform human tasks.
Zoom to the present day in 2021 and AI is a pervasive term in our day-to-day business lives, overused (many times incorrectly) and misunderstood by many. A key first step in evaluating AI solutions and vendor offerings is to level-set on a definition of AI. Ask the question “How do you define AI?” Is it the simplest version of a machine performing human tasks? Is it predictive? Or, are we talking about Skynet and self-awareness? (Terminator reference and a big fan). Establishing this mutual definition will help in all follow-up conversations and help avoid improperly set expectations.
Why Do You Use AI?
Using AI for the buzzword-effect’s sake can be a waste of precious resources and provide little to no advantage if used in the wrong circumstance. AI should be used in situations where intense human involvement is a drain on the organization, or where the consumption and analysis of large volumes of data can provide competitive advantage. Essentially, driving enhanced business outcomes should be the primary use case.
Ask the question, “Why are you using AI?” Any software vendor should be able to outline the advantage provided over non-AI means and clearly state the enhanced business outcomes and resultant efficiencies.
Where do you use AI and where is it hosted?
Understanding where in the process or business solution AI is used can be very telling and give you a feel for your vendors comprehension of the requirements and expected business outcome. Does the model align with your understanding of the pain point? Will it actually alleviate delays and errors and provide enhanced levels of productivity? Ensuring your hands-on business teams validate the Where is important as it helps guide your project and results.
The second Where question focuses on an important question: where does your AI live? Obviously, the rise of the cloud has made AI infrastructure a no-brainer, with cloud companies like AWS and Microsoft providing application developers and data science teams access to unlimited horsepower and key components for rapid development and management. Is a cloud environment right for your organization or do you need a containerized offering that can run on-premises?
How do you use AI?
The final How essentially ties all the previous questions and answers together. I’ll give an example. You don’t hit a tack with a sledgehammer, and not all business problems require AI as a solution. Here at Ephesoft, we leverage our AI in many of the complex document tasks where it provides excellent accuracy and business outcomes in both of our platforms. For example, we have a patent pending on our Semantik neural network for identifying and extracting complex invoice table data, which uses AI and supervised machine learning technology. However, not every problem requires that type of solution or the overhead of AI processes. For example, if you need the system to classify high-volumes of document types, AI may be too slow for the desired business outcome and may not meet your service level agreement because in this case, there are non-AI processing techniques that are much faster. The How needs to make sense, the right tool for the right job.
Hopefully, this article provides some insight into how to discern real AI from fluff and misconceptions so your next digital transformation project can achieve your expected business outcomes. If you would like to discuss further, contact us today.
