Technology

At FSTEC, a glimpse of restaurants' AI-powered future

Operators big and small testified to the transformative benefits of the technology. “I think the restaurant experience is going to look much different,” one attendee said.
Leaders from Chick-fil-A and McDonald's shared their AI strategies at FSTEC. | Photo by W. Scott Mitchell Photography

If your restaurant isn’t thinking about artificial intelligence yet, it probably should be.

That was perhaps the biggest takeaway from this year’s FSTEC conference, where AI dominated conversation both on the stage and off. The restaurant tech conference and trade show was held Monday through Wednesday at the Gaylord Texan Resort outside of Dallas. 

And while it would be easy to label AI the week’s biggest buzzword, it was clear by Wednesday that the technology is poised to have staying power in the restaurant industry. 

The tone was set early by keynote speaker Mike Walsh, a futurist who specializes in preparing businesses for the next big thing. He argued that we’re in the midst of a another industrial revolution—one powered by artificial intelligence rather than steam or electricity. “AI will not only change the way we live, it’ll change the way we work,” he predicted.

He was speaking in the future tense (he’s a futurist, after all). But operators at the event testified that AI is already having significant impacts on their business. To name just a few examples:

  • Yum Brands’ AI-powered inventory management system has reduced out-of-stocks by 90%. 
  • AI has improved the accuracy of Chili’s sales forecasts by 20%, allowing the chain to save more than 600 labor hours a week nationally.
  • Antares, a Burger King operator in New Zealand, is seeing 55% of customers upgrade their meal thanks to a nudge from Patty, an AI drive-thru voicebot with upselling capabilities.

“We’ve been able to get some really good ROI just from Patty sitting in the drive-thru doing his or her thing,” said Stephen Brown, Antares’ head of operations excellence.

“If we are intentional, we hope we can achieve a 15% productivity gain.” —Catherine Roberts, Chick-fil-A

With results like that, it’s no surprise that some of the very biggest restaurant companies are making big investments in AI.

At burger giant McDonald’s, employees can now turn to a chatbot called Ask Pickles when they need help troubleshooting a problem, like a broken soft-serve machine, said Wei Manfredi, McDonald’s VP of global architecture, data and generative AI. And the chain is still working on putting AI in the drive-thru, she added.

According to Yum Chief Digital and Technology Officer Joe Park, the owner of Taco Bell, KFC, Pizza Hut and Habit Burger developed a proprietary platform called Crave AI that analyzes data from all of its restaurants and then uses it to help them perform better.

Chick-fil-A, meanwhile, created an ethics framework around generative AI and even has a program for naming chatbots, said Catherine Roberts, director of intelligent automation capabilities. The chain hopes that the conscious approach will yield big things. “If we are intentional, we hope we can achieve a 15% productivity gain” among support center staff, Roberts said. 

While the industry's titans tackle AI head-on, smaller operators are taking more of a point-by-point approach. 

“Don’t tackle it all at once,” said Jerry Phillips, VP of technology for Whataburger, which is using AI to help forecast sales. He recommended that restaurants pick one or two use cases for the technology and go from there. 

Before a brand can even do that, though, it has to have the right foundation in place. Restaurant data tends to be disparate, Phillips said; centralizing it is step one. 

“AI starts with exposing your data,” said Clinton Anderson, CEO of tech supplier Fourth. “If you don’t have clean access to your data, that’s the fuel that runs your engine.”

From there, AI can start making connections about what’s happening in stores and generating useful intel for operators.

“Think of it as an eighth grader. It’s learning. It’s not going to be perfect.” —Jerry Phillips, Whataburger

But not everyone is there yet. The CMO of a 100-unit burger chain said his brand needs a new POS system before it can think about investing in AI. The brand’s current tech stack is a patchwork of old solutions, which is not the best environment for AI to thrive.

And while the general sentiment on AI was enthusiastic, there were some skeptics in the crowd.

“AI is like teen sex,” said Olga Lopategui, founder of Restaurant Loyalty Specialists, quoting a maxim from LinkedIn. “Everybody talks about it, nobody knows how to do it, and when it happens, everyone thinks it should have been better.” 

Sean Thompson, the VP of IT at Freddy’s Frozen Custard and Steakburgers, is not sold on using AI voicebots in the drive-thru, for instance. He pointed to the bots' order accuracy rate of around 93% and said that if a drive-thru headset failed  7% of the time, he’d get new headsets.

“We’ve all seen the viral videos that go out when these things go horribly wrong,” he said. “So we threw that out and said that’s not what we’re gonna be doing.”

That said, Thompson is not anti-AI. In fact, he was part of a panel discussing its virtues. But he was also realistic about its limitations. For one, it's not cheap, which means there has to be a clear return on investment. “Without ROI, you’re lost,” he said. 

Still, most attendees seemed to agree that AI will get better over time. “Think of it as an eighth grader,” said Phillips of Whataburger. “It’s learning. It’s not going to be perfect.” 

The more data it consumes, the better it gets. Yum Brands’ AI-powered inventory system took a year to launch at KFC, but just six months at Taco Bell. The company expects the timeline to be halved again at Pizza Hut. Park called it a “snowball effect.”  

“I think the restaurant experience is going to look much different.” —David Roberts, Cicis Pizza

A few times, operators were asked to imagine what restaurants will look like in five years if that snowball keeps growing at its current pace.

“I think the restaurant experience is going to look much different,” said David Roberts, CIO of Cicis Pizza. AI will handle some repetitive tasks, such as punching orders into the POS, and allow employees to “up-level” their skills in a way that elevates the brand, he predicted.

Almost every operator was insistent on that last point: AI won’t replace human restaurant workers, but will allow them to do other things that are more valuable. “You don’t do this to your employees, you’re doing this for them,” said Carl Nank, a restaurant operations specialist and advisor to AI voice supplier Hi Auto.  

The trick, he said, is figuring out what those freed-up employees should be doing, because that is where the return on investment lies. “ROI extraction is, what do I do instead, how do I engage the team to make them wanna do it, and then how do I make that start appearing to the customer as a value add that brings them back?”

The bottom line, though, is that some restaurant jobs are going to be automated away. 

“I think we need to be honest about that,” said Thompson. But he also pointed out that at many restaurants, employees are currently being asked to do too much. “Let’s take a few hats off of people and let them do what they do really, really well,” he said. 

At Bojangles, a drive-thru voice bot called Bo-Linda has allowed the chain to cut some hours, which helps the chain cover the costs of the technology, said CIO Richard Del Valle. But, “a lot of the hours we’re shaving out of the system are Fantasy Employee No. 2 that we don’t have in the first place,” he said.

“AI won’t replace humans, but humans using AI will replace humans.” —Joe Park, Yum Brands

Ultimately, there was a strong sense that humans will continue to be at the heart of restaurants’ operating model.

“Our customers enjoy our team members’ presence, hospitality and interacting with them,” said Park of Yum Brands. “Human touch is just something that is innate to the human condition.” 

AI won’t replace humans, he added, “but humans using AI will replace humans.” 

If there was one reason for pause aired at the event, it was that restaurants could be moving too quickly with AI.

“Everybody is trying to build this right now,” said Phillips of Whataburger. “I worry a little bit about the governance” of the data.

Roberts of Cicis also used the G-word. “I sure hope that everybody spends enough time with IT governance,” he said.

Some companies are taking steps to do just that. Little Caesars Enterprises, for instance, created an AI task force to develop guidelines and understand where AI exists in its system and what it is doing.

“That’s a best practice,” said Chief Information Security Officer Afia Phillips, “mainly to make sure you have that governance in place.”

Members help make our journalism possible. Become a Restaurant Business member today and unlock exclusive benefits, including unlimited access to all of our content. Sign up here.

Multimedia

Exclusive Content

Leadership

Cracker Barrel's CEO sees a long road ahead, but with returns already emerging

The 55-year-old brand expects to spend its next fiscal year testing what the concept should become to benefit guests, employees and shareholders alike, CEO Julie Masino said.

Financing

Gas prices are falling. So what does that mean for restaurants?

The Bottom Line: The price of gas has been falling much of the year and is below $3 per gallon in some states. Will that give operators a sales boost?

Financing

In Starbucks, an example of technology gone too far

The Bottom Line: As chains add more and more technology to get customers through the door, they may want to look at the issues at the coffee shop giant.

Trending

More from our partners