IBM Watson is going from winning “Jeopardy!” and diagnosing illnesses to predicting retail shopping trends. (Photo by Andrew Spear for The Washington Post.)

The buzz around consumer-grade artificial intelligence continues to grow. The latest example is IBM Watson Trend, a new way for consumers to understand the top shopping trends of the holiday season. Using a mix of machine learning, sentiment analysis, keyword analytics and natural language analysis, IBM Watson Trend tracks the top 100 selling items across three popular categories: consumer electronics, toys and health & fitness.

On the surface, most of what IBM Watson Trend is reporting as a “trend” is relatively obvious and probably doesn’t require a supercomputer. For example, the top three products in the consumer electronics category so far are the Apple Watch, Samsung TVs and Sony TVs. (The Samsung Gear VR, alas, didn’t even show up in the top 10, despite selling out on Amazon and Best Buy.) In the “toys” category, where the three leading trends are Star Wars LEGOs, LEGO Friends, and LEGO City, there’s also not much new.

This leads to the inevitable question: How do the shopping insights from IBM Watson differ from what’s already available on the market? By now, you’ve probably read more than a few articles about what products are set to fly off the shelves starting this Black Friday. And, yes, many of them inevitably mention the Apple Watch or “Star Wars.” So what’s the big deal about using AI to predict shopping trends?

According to IBM, there are two key advantages to putting the supercomputing resources of IBM Watson to work on analyzing holiday shopping patterns. One is that IBM Watson is able to crunch a lot more data than if you’re not using a supercomputer – tens of millions of online conversations and 10,000 sources and counting. Think of IBM Watson as a voracious online reader, able to scan social networks and media publications in real-time for any sign of a change in consumer sentiment. It’s then able to assign a “Trend Score” on a scale of 0-100 to give you a sense of how confident this trend is real (and not just a fad).

Another key advantage is that IBM Watson, by using natural language analysis, can parse a lot of online chatter that’s notoriously tricky to decipher. On the Internet, remember, grammar, diction and spelling rules are often optional at best. People like to use emoticons rather than words these days. And sarcasm and irony are difficult to get right. Even something as obvious as, “I’m SO going to buy this” could be interpreted in various ways, depending on context and tone of voice.

For a hint of how consumer-grade AI could change the way we shop, it’s interesting to see what IBM Watson has uncovered that’s surprising or unexpected. Nikon cameras, which ranked No. 7 in the consumer electronics category, might seem obvious until you read IBM Watson’s back story on them – what IBM refers to as “the story behind the trend.” It turns out that smartphone users who are also amateur photographers are among the biggest potential buyers of Nikon cameras. They’ve been so turned on by the potential of smartphone photography (think: Apple’s “shot on iPhone 6” campaign) that they want to upgrade:

“The Nikon camera trend is driven by conversations related to Nikon DSLRs being a natural upgrade from smartphone cameras,” says IBM Watson Trend. “The conversation around Nikon cameras tends toward people who are just getting into higher-quality photography.” In contrast, expert photographers, turned on by the allure of high-definition Internet video, are turning to Sony’s new 42 megapixel Alpha 7R II camera, which boasts the ability to record Ultra HD 4k video.

It’s not so much that IBM Watson Trend is able to predict the future, then, it’s that it gives people a useful tool for interpreting available data in new ways. IBM says that Watson can help to pinpoint patterns and trends, based on a combination of content, context and sentiment. Right now, the only trend that IBM Watson is 100 percent convinced of is the Apple Watch, which continues to be among the most discussed products.

Another example of AI at work is how IBM Watson uses information about the upcoming new “Star Wars” movie to analyze how it might impact the toy market. It’s a no-brainer that “Star Wars” toys are going to be big this year – and Star Wars LEGOs are currently the No. 1 top-ranked toy — but it’s less obvious how “Star Wars” is going to impact other toy-buying behavior: Will these LEGOs “crowd out” certain toys? Or will they give a “halo effect” to similar products?

IBM Watson thinks that the popularity of “Star Wars” LEGOs is going to have a halo effect for other LEGO products, particularly LEGO Friends (ranked as the No. 2 toy) and LEGO City (ranked No. 3). The logic is simple: parents walk into the LEGO Store, see a “Star Wars” LEGO set and decide to add on a few more products on their way out the door.

From this perspective, IBM Watson Trend appears to be more of a business retail solution than a consumer solution. It makes retailers smarter about what to stock, and how to pitch products to consumers. At this year’s IBM Watson ecosystem event in Brooklyn, for example, there were a number of IBM Watson partners trying to crack the code of e-commerce by leveraging the power of cognitive computing. If you know exactly what a consumer has in mind when he or she visits your Web site, you can make the sale much more easily than if you have to ask a lot of questions.

What will be interesting this holiday season is to see how well IBM Watson handles the gap between “aspiration” and “intent.” While (almost) all of us may aspire to own the new Apple Watch and may relentlessly hype it up to our friends on social media, how many of us are actually going to plunk down the cash and buy one? As anyone who has ever tried to predict anything knows, life rarely turns out exactly the way we expect.