When technology gets too clever for its own good: Why 'Mitchell' is now a dirty word according to this fast food chain’s computer

There's nothing quite like the feeling of being told off by a machine, especially when you're just trying to order some chicken and chips.

That's exactly what happened to Mitchell from Sydney when he popped into his local KFC for a quick bite, only to have the self-service kiosk essentially wash his mouth out with soap.

The trouble started when Mitchell typed his perfectly ordinary name into the ordering system, only to be met with a stern digital rebuke: 'Please try again, the word is not allowed.'

His crime? Having the audacity to be named Mitchell - apparently because his name contains the word 'h*ll.'

What Mitchell experienced has a name, and it's been plaguing the internet for decades.

Image source: Reddit.

It's called the 'Scunthorpe problem,' named after the English town that found itself blocked by overzealous content filters because it contains an unfortunate four-letter word.

The problem occurs when automated systems designed to block offensive content cast their net so wide they catch innocent words, names, and phrases.

Recent examples include social media’s moderation system flagging posts mentioning 'Luigi' (thanks to Luigi Mangione making headlines), and even the US Department of Defense mistakenly removing references to the historic Enola Gay aircraft.

While younger tech users might laugh off these digital hiccups, research shows older adults face particular challenges when services convert to digital systems, especially when self-ordering kiosks are made to seem familiar but don't behave as expected.

The 'digital divide' affects many seniors who struggle with accessing and using modern devices due to limited familiarity with technology, and while some have adapted, others face real difficulties navigating apps and digital tools.

Mitchell's experience resonates because it highlights how these systems can make a simple task—ordering food—unnecessarily complicated and embarrassing.

The issue stems from using basic 'substring matching'—the oldest and most straightforward way to do automated text moderation using lists of keywords related to potentially problematic content.

As one social media user astutely observed about Mitchell's situation, 'It's just using basic substring matching of characters. Whoever got the job to program this lied on their resume.'

More sophisticated systems use language models that consider context and culture, understanding that the same word can appear in different contexts—some problematic, others completely innocent—but many businesses still rely on the cheaper, cruder approach.

KFC itself has been struggling with customer satisfaction, experiencing monthly traffic declines of 2-12 per cent every quarter in 2024 and 2025, with overall sales down 5 per cent in 2024 and a further 4 per cent decline in early 2025.

While self-service technology is meant to improve efficiency, incidents like Mitchell's suggest it might be driving customers away.

The problem extends far beyond fast food. Social media platforms like TikTok have laid off 700 human moderators in favour of AI systems, creating what experts call a 'fighting fire with fire' mentality, using AI to moderate AI-generated content.

Read more: Today’s receipts are twice as long...but half as useful! Here's what's missing

Have you had your own run-in with an overzealous computer system? We'd love to hear about your experiences—and any clever workarounds you've discovered—in the comments below!