No one knows what a head of AI does, but it’s the hottest new job

America’s biggest companies are hiring AI leadership as fast as they can.

If AI is coming for our jobs, many Americans are hoping to get out in front of it. Regular people are using AI at work, and tech workers are rebranding themselves as AI experts. And those in leadership are vying for the hottest new job title: head of AI.

Outside of tech, the head of AI position was mostly nonexistent a few years ago, but now people are taking on that title — or at least its duties — at everywhere from Amazon to Visa to Coca-Cola. In the US, the number of people in AI leadership roles has grown threefold in the past five years, according to data from LinkedIn, bucking the downward trend in tech hiring overall. And while the head of AI job description varies widely by company, the hope is that those who end up with this new responsibility will do everything from incorporating AI into businesses’ products to getting employees up to speed on how to use AI in their jobs. Companies want the new role to keep them at the forefront of their industries amid AI disruption, or at least keep them from being left behind.

“This is the biggest deal of the decade, and it’s ridiculously overhyped,” said Peter Krensky, a director and analyst at Gartner who specializes in AI talent management.

Like anything new in tech, the AI revolution can take on a bit of a gold-rush quality. AI is one of the few areas where companies are actively spending money, since they see it as the inevitable future and as a way to improve their bottom line. At the same time, the parameters of the head of AI job — and even AI itself — aren’t very clear, and the pivot to the position can seem opportunistic. Remember Shingy, AOL’s Digital Prophet?

The thing is that while everyone seems to agree that companies need AI stewards, the nature of new technology means many are uncertain as to what that stewardship means in practice. Furthermore, we’re not sure about who exactly should become the new stewards: the people who have been working on AI for years or those who have been introduced to AI by the latest crop of consumer products and understand how the rest of us use it. We’re also not certain just how big of a disruption AI will be and how fast that disruption will happen.

Those are just some of the reasons companies are hiring heads of AI. And if they don’t already have a head of AI, most big companies will have one soon.

“If I were talking to a CEO a year ago, and I was like, ‘You’d be a fool not to have a head of AI.’ They’d be like, ‘Come on, give me a break,’” said Krensky. “And now they’re like, ‘I know, that’s why I have one.’”

Krensky estimates that currently about a quarter of Fortune 2000 companies have dedicated AI leadership at the VP level or above. He expects it to be about 80 percent a year from now. While the position will be more commonplace at bigger companies — especially those in banking, tech, and manufacturing — he’s also seeing it crop up at midsize organizations and in government agencies.

Typically, the person taking what Kensky calls a “cool and sexy” job title — one that he says is often a “hat, not a role” — comes from an existing technology leadership position like chief data officer or chief information officer. But the accessible nature of generative AI tools and their potential use across industries and positions has meant that people in nontech roles like business and marketing are also donning the mantle.

And because AI is supposed to be more transformational and more readily profitable than tech fads like Web3, experts think the head of AI is also going to stick.

“This is going to be a role that will stay on for a while. It’s not a transitional role,” said Beena Ammanath, executive director of the Deloitte AI Institute. “It’s absolutely crucial.”

Just what any given head of AI does varies, especially depending on the type of company. Generally, that breaks down into heads of AI at digital companies working to incorporate the technology into their products, while at nontech companies that means figuring out where and how to use existing AI technology to improve their business models. Everyone, it seems, is trying to get the rest of their company to start using AI.

Mike Haley, SVP of research at Autodesk, says he’s the company’s de facto head of AI, having guided the architecture and engineering software company’s AI strategy for more than a decade. In addition to steering AI usage within the company, Haley is invested in putting AI to use in Autodesk’s products in order to “dissolve the interface” between users and the software. That means AI could help people use “natural methods of expression” like English or a pencil drawing, for example, to create detailed blueprints.

“Suddenly this complex tool that requires all sorts of learning and parameterization becomes way more accessible to more people,” explained Haley, who has a background in computer science and applied math.

Bali D.R., head of AI and automation at IT services consulting firm Infosys, is helping clients leverage AI while also trying to use it to “amplify human potential” across Infosys, from recruitment to sales to software development.

“All parts of the value chain, we are seeing how we can actually make it better, faster, cheaper,” says D.R., who moved to the AI role from another management role, and who started his career at the company 30 years ago in software development.

FICO chief analytics officer Scott Zoldi has been leading the data analytics company’s AI efforts for the past seven years, also without the “buzzy” head of AI title. He’s mainly focused on incorporating AI into the company’s products, including using consumer spending patterns to help detect credit fraud or when a customer is falling for a scam. He also spends a lot of time thinking about how AI can be used responsibly so as not to run afoul of regulatory bodies, corporate governance, or consumers by, for example, using AI that’s more likely to flag a certain group of people for committing fraud.

Zoldi, who says he’s written more than 100 AI patents, thinks the “head of AI” position should go to someone with a technology background.

“You really have to be an expert or you’re potentially going to be setting up the organization for failures down the road because it’s very complicated,” Zoldi, who views the position as a sort of watchdog, like a chief of security.

While Gartner’s Krensky estimates about 80 percent of AI leadership comes from a tech background, another 20 percent, of course, does not.

That’s the case with Coca-Cola’s global head of generative AI, Pratik Thakar, who previously led the company’s global creative strategy.

Pratik has been using AI to streamline and amplify the company’s advertising products. That included recently using AI to make roughly 15 percent of a commercial, which sliced the production time from a year down to two months.

Conor Grennan, a dean at NYU’s Stern business school, who recently took on the additional title of head of generative AI, sees the title as more of an initiative and thinks of it as akin to a chief learning officer or chief productivity officer. In the position, he pushes people across NYU, from students to professors to administrators and recruiters, to use AI to become more efficient and better at their tasks.

Grennan, who has an MBA and had previously studied English and politics, thinks it’s actually better for many organizations if their AI leadership doesn’t come from a tech background so that the person is better able to explain its benefits to a wider audience of mere mortals.

“You don’t need to know the software running your iPhone, just order an Uber,” Grennan said. Instead, what’s important for the role, he says, is creativity with language and breadth.

“They need to be an excellent communicator, they need to have a view of the entire firm, at least at the 30,000-foot view. And also it has to be somebody who really understands what generative AI can do,” Grennan said. “You don’t capture everything by putting it in the tech department.”

Regardless of where the head of AI sits within an organization, the fact remains that it’s a new frontier that will likely change a lot as the technology and our understanding of it develops. And like with any new technology, there’s going to be a mix of genuine innovation and genuine swindling.

AI is happening, and it will be a very big deal. But its full effects — and exactly what those are — will roll out over many years, so we may have time to figure things out.