AutoZone Worker Claims Product Is Sold Out. Then The Customer Pulls An Uno Reverse: 'We Are Cooked'

AI has made its way into our everyday lives, from figuring out what to eat to what to say. It’s gotten to the point where people trust AI so much that they'll use it to second-guess human experts.

Those working customer-facing jobs now have to deal not only with people who think they know more than they do, but also with people further emboldened by unverified claims from AI.

Does ChatGPT Know AutoZone’s Inventory?

In a TikTok with more than 182,000 views, content creator Mya (@bluesclovers) shared a moment that apparently left her questioning the future of customer service interactions.

In the video, Mya told a customer that AutoZone doesn't carry a particular product. She thought the interaction would be done there since the customer had left, but a few minutes later, he came back, insisting she was wrong.

"He came back inside a few minutes later to tell me that 'AI told him we DO stock it,'" she said in the text overlay.

"We are so cooked,” she added.

"Last I checked, ChatGPT doesn't work at AutoZone,” Mya went on to say in the caption.

What Is An AI Hallucination?

Here's the thing about AI chatbots—they don't actually know anything.

They're not searching a database or checking live inventory. According to IBM, what they're doing is predicting what a plausible-sounding answer looks like based on patterns in the data they were trained on.

Sometimes that lands on something accurate. Sometimes it just confidently makes something up. That's what's called an AI hallucination. When the model produces something that sounds totally reasonable but is wrong.

It's happened at scale in pretty memorable ways. Google's Bard once told users the James Webb Space Telescope had captured the first-ever images of a planet outside our solar system. It hadn't. Meta's Galactica model got pulled after three days because it kept generating inaccurate, sometimes biased information, all delivered in the same matter-of-fact tone as the stuff it got right.

The issue is AI doesn't flag its own mistakes and it sounds equally sure of itself whether it's correct or not, which makes it a bad source for anything that requires real-time information.

How Do People Feel About AI?

According to Pew Research Center, 31% of Americans now interact with AI at least several times a day, up from 22% in early 2024. Public awareness has surged too, with 47% saying they've heard a lot about AI, a 21-percentage-point jump since 2022.

About one in five US workers now say at least some of their work involves AI, up from 16% the year prior. Younger adults are driving the adoption: around half of adults under 50 say they interact with AI at least once a day, compared to significantly smaller shares of those 50 and older.

But familiarity hasn't translated into confidence. Half of US adults say the increased use of AI in daily life makes them more concerned than excited, up from 37% when Pew first asked the question in 2021.

Only 10% say they're more excited than concerned. Americans tend to be more open to AI in data-heavy tasks like weather forecasting, but about half say AI will worsen people's ability to think creatively or form meaningful relationships.

ChatGPT Doesn’t Know As Much As You Think

This experience seems to be becoming more common with people thinking that with AI behind them, they have a harbinger of truth in their back pocket, even though AI can be very, very wrong.

“Do they...do they think ChatGPT is some omnipotent harbinger of the truth?” one wrote.

“‘ChatGPT isn't a real person and therefore knows nothing, it compiles every random thing on the internet and spits it out into what it thinks sounds right’ is my fave response to that bs,” a person said.

“Had the same interaction except it was for a book that was out of print and we could not order. ‘AI overview says you have it in stock and no offense but it’s a lot smarter than you and I’ LIKE WE ARE SO COOKED,” another wrote.

“I work in a paint department and everyone’s using ChatGPT generated color and number names that don’t exist,” a commenter shared.

Motor1 reached out to Mya for comment via TikTok direct message and comment and to AutoZone via email. We'll be sure to update this if they respond.