What Marketers Are (and Aren't Yet) Doing With Artificial Intelligence

Artificial Intelligence is here, yes, but it’s important to remember that the current capabilities of AI are a small part of what it can and probably will look like in the future, according to a panel gathered today for Adweek’s Elevate: AI summit. “This tech is still in its infancy,” Conversable and Hypergiant CEO Ben Lamm said he likes to remind his clients. Sherif Mityas, chief information officer and chief strategy officer at TGI Fridays, used that same metaphor when describing AI at his organization. “We code-name our AI products with the latest celebrity baby names because they are still infants,” he said. And the sentiment was echoed by other panelists as well. While most are actively incorporating AI and machine learning into their work, they’re also constantly testing, reevaluating and planning for a future when the technology will play an even bigger and potentially very different role within their organizations. “We’re on a mission to become the most creative and impactful data-led organization in the world,” said Kimberly Kadlec, Visa’s svp of global marketing. “We’re going to move to a model where we have a very, very high percentage of precision marketing and automated marketing." But achieving that requires a lot of experimentation and a certain level of comfort with failure. “You have to be comfortable with telling your boss, management, ‘We tested it out; it’s wrong,’” Kadlec said. “Otherwise, people are too afraid to push the boundaries.” It’s a similar approach that Lamm offers his clients, advising them to “start with something very small and learn, grown, iterate.” As a young technology, AI certainly has its limitations, but it also makes it rife with possibility. While AI can help organizations figure out how to speak to audiences, it can also let organizations figure out whom not to speak to, helping clients discern “people likely to unsubscribe and being careful about what to send them,” said Chris Jacob, director of product marketing for Salesforce’s Marketing Cloud. Effective internal communication is also important. “There’s marketing speak and tech speak, and they do not meet," Kadlec said. "We’re creating a language that both teams can understand." The uncertainty of what AI will be when it grows up can be exciting, but there is another, more uncomfortable uncertainty to address, especially as the pace of AI integration increases: the displacement of workers by tech. For now, though, personnel issues around AI tend to go in the other direction. “Demand is far outstripping the supply,” Jacob said of the pool of talent that can effectively employ the technology. Rather than merely waiting for a new crop of AI specialists to make its way through Ph.D. programs, Jacob thinks it’s important for organizations to take responsibility for employee education, which he said "has to be the underpinning to everything." For Mityas, the challenge is in ensuring human and machine keep a comfortable pace with each other. “How do you lockstep the organization and the people with the tech so one doesn’t get too far ahead of the other?” he said.