Recently, under the auspices of AMA-New York's Catalyst for eCommerce Series—co-sponsored by Channable—four experts gathered to give their best shot at answering it. The event took a long-range look at the future of AI in eCommerce, tackling the moment's most-hyped subject from a variety of angles.

Read on for five key takeaways on eCommerce's bright, shining, somewhat scary, and comprehensively AI-facilitated future.

1) Generative AI is a very different beast from legacy AI

The first thing it's important to note is that marketers have been using AI in eCommerce for at least a decade. As Hunter Hartwell, Partner at Glasswing Ventures, pointed out, this legacy AI was "deterministic in nature."

Marketers would turn to a machine algorithm to solve a problem with a concrete, objective answer—say, the best way to disperse ad inventory or position a banner ad.

What makes generative AI "earth-shattering," per Hartwell, is its ability to generate new content whole-cloth—to take in pre-existing material and spit out something that didn't previously exist, whether that's advertising copy or a quarterly report. It's this distinction that is at the root of all the exciting developments in AI-facilitated eCommerce right now.

2) If you think you know what "personalization" means in an eCommerce context, think again

Right now, in the dying days of the third-party cookie, personalization in an eCommerce context takes on a few familiar forms—say, the sunglasses you look at once which then follow you around the internet for a month.

These techniques aren't going anywhere anytime soon, but the level of personalization facilitated by AI in eCommerce—as sketched out by the panelists—does suddenly make them look old-fashioned.

Per Cara Schlanger, current freelance consultant and former Senior Vice President of Global Marketing & eCommerce at Williams-Sonoma, "hyper-localization and hyper-personalization" are the future. Already, she pointed out, there are lifestyle companies offering radically personalized experiences, like skincare brands that can analyze your skin type and then recommend the best product, or home decor brands that let you test out products in a virtual version of your own living room.

But that has nothing on what's coming next—what Schlanger describes as the "personalized upsell."

In the not-too-distant future, according to Schlanger, AI will have the ability to analyze a customer's shopping cart and (if available) their past preferences and then actually invent a customized product on the spot to sell to them. Kapil (Kap) Sharma, Technical Lead at Wipro, even floated the idea of personalized influencers designed specifically for individuals or households—fake people spun up via AI to get consumers on board with a brand.

3) Marketers aren't getting replaced by robots any time soon—but AI in eCommerce will change their job responsibilities drastically

It's true, as Schlanger pointed out, that "some disciplines [within marketing] may disappear" in the near future. At the same time, though, Sharma says, "with AI the power of the marketer goes up quite a bit.”

Marketers can be marketers rather than data jockeys—that's exciting to me," Schlanger said.

In other words, instead of wasting time on repetitive manual tasks—like bugging the guys in the data science department to pull specific reports—marketers can quickly generate that information themselves and put it to creative use. "A single marketer can now wear several hats," said Hartwell, rapidly generating campaigns and reports and ad copy, A/B testing on the fly, compressing weeks of trial-and-error into a playful afternoon.

4) AI in eCommerce might soon change almost everything—but no matter what happens, brands will always matter

All panelists were at pains to point out that AI will always just be a co-pilot: brands will always matter, and in fact they might matter even more in a world of AI-facilitated deep-fakes and unreliable, fly-by-night drop-shipping companies.

Schlanger brought up the experience of finding a too-good-to-be-true product on Instagram and then—when it arrives in the mail—promptly discovering that it actually was too good to be true. "Brands continue to matter," she said. "AI just allows us to move faster and smarter and to better-understand what the customer wants."

In the context of AI in eCommerce, humans will likely always be at the wheel, shaping the inputs, and the brand voice. Now, they'll just be navigating with a much more powerful vehicle.

5) Setting up in new markets will soon be a breeze

Launching a brand in a new market is currently an involved ordeal, requiring not just language skills but also a deep awareness of what resonates with a given population. AI is set to turbocharge that process by enabling companies to "go global faster," in Schlanger's words.

As Hartwell pointed out, using AI in eCommerce will allow marketers to translate the copy and tweak the product design "so it fits different trends in different countries." Instead of laboriously rolling out a brand market-by-market over a period of years, brands can effectively hit a few buttons and at least experiment with other markets—even if it doesn't work, AI lowers the cost (and thus the stakes) of brand expansion.

In fact, Channable is already allowing brands to go to market faster and gain global visibility in record time.

By tapping into the power of AI, it lets brands categorize millions of products while connecting them to thousands of new markets, at a speed that would’ve been impossible just a few years ago.