McDonald’s Halts AI Drive-Thru Tech: What’s Next for Fast Food?

McDonald’s is pulling the plug on artificial intelligence ordering technology it was testing at more than 100 restaurant drive-thru systems in the US, a possible hiccup in the rapid rollout of AI in the fast food industry.

The company had worked with IBM to develop and test AI-driven, voice-automated ordering at some of its restaurants. The fast food giant says it’s still working on AI-related solutions at a time when its rivals are making similar investments.

It plans to shut off the technology in restaurants participating in the test “no later than July 26, 2024,” according to reporting from trade publication Restaurant Business.

“IBM remains a trusted partner and we will still utilise many of their products across our global System,” McDonald’s said in an email sent to franchisees. But the company suggested it will look at AI partners other than IBM.

McDonald’s and IBM launched their partnership in 2021, announcing the development of Automated Order Taking (AOT) technology to create a more convenient and simplified ordering experience for its customers and restaurant teams as a part of its “Accelerating the Arches” growth plan.

IBM said the AOT technology that emerged from the partnership has “some of the most comprehensive capabilities in the industry, fast and accurate” in demanding conditions.

“While McDonald’s is evaluating and refining its plans for AOT, we look forward to continuing to work with them on a variety of other projects,” IBM said in a statement.

McDonald’s says this is not the end of its AI efforts. It plans to “evaluate long-term, scalable solutions” forvoice-ordering by the end of 2024.

McDonald’s is not the only fast-food chain experimenting with AI. Other companies like White Castle tested out an automated drive-thru ordering system in 2021, and Wendy’s expanded its partnership with Google Cloud to roll out its own AI ordering tool in May last year.

Despite fast-food companies’ enthusiasm to integrate AI-powered solutions into their daily operations, the technology has still hit snags. Some customers have complained about AI getting their orders wrong — a result of the technology’s inability to recognize some accents and distinguish the customer’s voice from background noise.