The rise of the chief automation officer

Whether it’s labeled chief automation officer, chief AI officer or chief innovation officer, the buzz about a new C-suite position reflects the current fixation on all things AI. And whether it’s just temporary hype or a real revolution in how companies are run, the talk about a chief automation officer title is just part of a broader reckoning with AI’s potential impact on the business world.

“At the bank, we say that innovation is everyone’s job,” said Bank of America’s David Boast, who led a discussion with three leading tech executives at the third annual Bank of America Triangle Innovation Summit. “Positions like chief automation officer — those roles have been around a little while but are being discussed more while we're in the middle of a hype cycle. So, is that a real role? Is it necessary? Is it going to last?”

Avaneesh Marwaha, CEO of legal technology specialist Litera, believes a C-suite AI position might be overkill. “I'm finding AI enablement living inside the HR team,” he said. “This is really about people transformation, right? We're changing jobs. We're changing how folks work, and that level of change and execution, day in, day out — it should sit with the people team.”

Meanwhile, Drew Peters, CFO of ServiceTrade, shared that capitalizing on AI’s potential requires leadership, and not necessarily just in the C-suite. “We do have a person — who is focused inwardly and reports directly to our CEO — who is tasked with driving AI through the business. But one person can’t do it all,” he said. “So, this year we've identified an AI champion in each department to make sure that AI permeates throughout the business.”

Those two points of view in some ways reflect two emerging roles for AI: first, reshaping core internal business functions and second, opening broader possibilities, including external applications.

For many companies, AI already is restructuring work by automating routine tasks and speeding up decision-making. In the CFO’s office, that might be forecasting or risk modeling; in operations, AI might help improve supply chain visibility and identify process improvements.

Within tech companies, early adoption was found in development, writing code — “the low-hanging fruit,” said Pendo co-founder Eric Boduch, now CEO at venture studio 24 And Up. Boduch said he’s talked with companies that increasingly are using AI to determine how human expertise is deployed in other departments; for instance, in customer service, the humans handle calls or client issues that can’t be simply solved by AI. “Those are probably the higher-value issues anyway, and you end up having a higher degree of expertise on those calls. I think it’s better for customers overall,” he said.

At ServiceTrade, Peters said, “We used to have a sales manager review maybe 20% of calls in a week and then try to identify how best to coach sales reps. Now, AI reviews 100% of calls and delivers personalized, actionable coaching directly to each rep, highlighting exactly where they can improve to be more effective. Managers still have visibility, but because coaching happens automatically and at scale, we get full coverage in far less time — freeing sales managers to focus on higher-value work while building more productive, higher-quality salespeople.”

Boduch and Marwaha said using AI to free managers from some of those routine responsibilities can — ironically — allow them to spend more time working face to face, whether that’s visiting clients or observing their staff in action.

Despite these successes, despite the hype and despite massive investments, many companies still complain that enterprise AI projects are not delivering ROI. Part of the problem is unrealistic expectations, combined with a lack of focus. “A lot of projects fail because people don't have a clear problem statement of what they're trying to solve — the baseline and how they're going to track against that to some form of success. What does success even look like in that case?” said Peters. “A lot of people just expect that because AI's here, it's going to solve all our problems. But for probably 95% of us, we don't really know how to use it to the greatest extent. So, it's not surprising that 95% of the time it's going to fail. Because if you just expect that AI is going to solve your problem, you're in for a rude awakening.”

Panelists said that at least for now, a large part of the savings AI will deliver will come directly from cutting manhours and reducing staff. Marwaha said Litera will see “a big reduction of our head count because of the AI investments we've made internally,” he said. “But we’re also looking at customer sentiment, usability, adoption and how AI can help make the customer journey better and frictionless — so measuring ROI is not always headcount related.”

Equating AI adoption with layoffs leaves some executives feeling squidgy. Said Marwaha: “When we go out to customers and do pilots, they actually start thinking about, ‘Wait — I might fire 10 people because of this project, right? And I could be one of the 10 that's going to be let go because you're automating parts of my job. You've given me so much value that, what am I going to do here?’”

But given the long-term potential, the panelists said it’s critical to keep pushing to find ways to move AI’s role forward, both for internal operations and external applications. For companies feeling that their AI initiatives are stuck, these executives suggested some approaches:

• Stop thinking about AI initiatives as pilot projects. “You’re not going to get the ROI out of a pilot project unless you turn them into real projects that are supported on an ongoing basis,” said Boduch.

• Hire help to figure it out. “You have a business to run. And now people say, ‘Hey, by the way, reexamine everything you’re doing in regard to AI to see where you can get more efficiencies,” said Boduch. “Figuring out how you can innovate with AI — it’s a challenge.” He suggested hiring a consultant to help.

• Task the AI experts with sharing expertise and enthusiasm. That includes top leadership. “It’s incumbent upon the folks that are the AI champions to help others along that curve,” Peters said.

• Learn to live with AI every day. Said Marwaha: “We're pretty firm on our belief that it has to be part of your day job every day. It's not something you just click once an hour. You should be living in the environment that we've selected for you to work in, because we think it's the best way to have great customer outcomes. It’s not because I want to squeeze every minute from your life. It’s because I want to train you to be competitive in the long run.”

• Provide dedicated time to learn, experiment — and fail. For Peters and ServiceTrade, that means “providing what we affectionately call ‘tinker time,’” he said. “But we also have to create an environment where there’s no fear of failure. I’m happy to try things—and fail fast.”