Sellers everywhere are reporting that their tools contain some form of artificial intelligence or machine learning. Here are three questions to ask to separate the technology from the marketing hype.
One of the myriad challenges of being a modern technology leader is separating marketing hype from reality when it comes time to buy new hardware or software. Product marketing often tends to exaggerate and focuses on the positive rather than the negative. With technology products, there is the added wrinkle of complex technical elements that require specialized understanding.
Combine the historical hyperventilation of most product marketing with hot technology, and you have to wallow through a dense wall of promises, buzzwords, and claims to determine if a product will work for your organization. This is especially true in the age of artificial intelligence, where everything from supply chain software to office furniture claims to have an element of AI embedded. You could almost imagine a late night infomercial host shouting that the product he’s shilling, “Now contains 30% more machine learning!”
TO SEE: Ethical Policy for Artificial Intelligence (Tech Republic Premium)
The problem with evaluating products that contain AI is that definitions of what AI is can vary widely. If your definition assumes learning algorithms that intelligently categorize new data, and your vendor believes that AI involves little more than a little imaginative computation, you’ll be disappointed. To determine what your supplier means when they tell you that there are elements of AI in their product, here are three simple questions that can help you separate the hype from the reality.
How does the AI model learn?
A fundamental element of most real AI technologies is that they improve based on the data they receive or incorporate technologies that test potential future results and amplify their calculations based on those results. Game-playing AIs are a classic example of this technology, where the AI can simulate playing thousands of iterations of a game and improve performance based on the outcome of each game.
Ask your supplier for some details about how the AI learns and improves. What data does it use? Does it simulate potential scenarios and use that to learn? How many simulations can it run? If you ask these kinds of questions, you may soon find out that the vendor touted “AI-driven learning” is really just a few basic calculations on your existing dataset rather than actually tweaking the algorithms based on a learning ability. .
How is the AI monitored and adjusted?
Actual AI systems adjust their predictions based on a combination of the input they receive and their ability to run various simulations to test possible outcomes. As such, the AI needs to be monitored and may need to be retrained or have additional input data.
Asking your supplier how the AI is monitored and modified will indicate whether their product actually has some degree of intelligence over some fancy standard algorithms. Suppose your supplier claims that no supervision or adjustment will ever be required. In that case, you can rest assured that AI is marketing hyperbole rather than an embedded and beneficial technology in the product in question.
Do you share customer data to train the AI?
Another critical question to ask your suppliers relates to two essential topics. First, it’s worth knowing if your data is mixed with other customers’ data to train the AI in a product. This may or may not be beneficial. For example, if you’re considering a supply chain management solution, letting your data inform the AI in exchange for the benefit of other companies’ data can be a valuable trade, as a more comprehensive dataset should make the product more effective. Conversely, if you are working with unique and very specific data, it can be a handicap to allow the AI to be influenced by other data.
This question should also lead to your supplier explaining how customer data informs and improves AI. Say they don’t have an answer to this question, or state that no customer data actually affects the AI’s ability to make predictions. In that case, it is a likely indicator that the product in question does not contain any real AI technology.
It’s easy to tout “AI Inside” as a benefit to a tech tool. However, the imprecise definition of artificial intelligence makes the job of a tech leader difficult. Using these questions to determine how far AI is powering your technology can be an important differentiator when selecting technology.
