Taco Bell scales back AI drive-thru after order errors and ‚creepy‘ reactions

Stylized editorial illustration of a sleek AI chip on a minimal studio background, soft gray and cobalt palette, matte metallic texture.

Taco Bell is dialing down its AI drive-thru experiment after reports of order mistakes and uneasy customer reactions, following a rollout to more than 500 locations in the U.S. The company found the voice assistant was error-prone, easily manipulated, and sometimes unsettling for customers, according to Gizmodo, which cites comments to the Wall Street Journal.

AI rollout meets real-world friction

Taco Bell’s Chief Digital and Technology Officer, Dane Mathews, told the Wall Street Journal the company is “learning a lot,” including that some customers intentionally tried to trip up the system with absurd orders like “18,000 cups of water.” Posts on Taco Bell’s Reddit community described incidents where the assistant allegedly told customers the restaurant was out of everything except drinks and sauce packets, and another where a request for a Chalupa Supreme with onions became three chalupas. In one case, an attempt to swap meat for beans was reportedly rejected by the system.

Despite these setbacks, Taco Bell’s parent company, Yum Brands, announced a partnership with Nvidia earlier this year aimed at improving the technology powering its AI operations, including automated order taking, Gizmodo notes.

Industrywide tests and mixed results

Fast-food chains across the U.S. have been testing similar systems. McDonald’s began introducing AI into operations to improve order accuracy, while Wendy’s partnered with Google to bring an AI chatbot to drive-thru windows and trained it on brand-specific shorthand like “JBC” for “junior bacon cheeseburger.” White Castle moved to expand AI to more than 100 drive-thrus via a collaboration with speech recognition company SoundHound.

Customer reactions and company reassessments

According to Gizmodo’s report, some of these initiatives have encountered familiar hurdles. McDonald’s reportedly pulled back on certain customer-facing AI after repeated order errors, and social media posts described Wendy’s AI as inaccurate or “creepy.”

Reflecting on Taco Bell’s own trial, Mathews told the Wall Street Journal that when restaurants are extremely busy with long lines, it may be preferable for human staff to handle orders. While the company is continuing to pursue AI improvements through its broader partnerships, the immediate lesson from the drive-thru pilot suggests the technology’s performance and customer comfort remain works in progress.

According to Gizmodo, similar experiments at other chains indicate the industry is still fine-tuning how and where AI can fit into fast-food ordering without compounding errors or frustrating customers.

Total
0
Shares
Pridaj komentár

Vaša e-mailová adresa nebude zverejnená. Vyžadované polia sú označené *

Previous Post
Isometric abstract nexus of interlocking nodes and circuit-like pathways in soft gray and cobalt, matte finish, no people

Nvidia’s $54B Outlook Revives AI Slowdown Fears as H20 China Sales Vanish

Next Post
Close-up technical illustration of a silicon microchip on a lab bench, soft gray and cobalt tones, matte metallic finish

Anthropic extends data retention to five years, offers training opt-outs

Related Posts