Hospitality businesses are facing a new wave of fraud driven by artificial intelligence, food safety experts have warned.
According to Food Alert, customers are increasingly using AI-generated images and legally framed complaints to secure refunds and compensation.
The company is already seeing live cases across the UK involving manipulated food images designed to depict mould, undercooked meat and foreign objects that were never present in the original meal, and has warned this is creating growing financial and reputational risks for operators.
It said customers are also increasingly using AI tools to generate convincing complaint emails that include legal terminology and references to enforcement authorities, in an attempt to pressure businesses into issuing refunds.
As a result, the food safety and compliance experts are urging hospitality businesses to strengthen complaint investigation procedures, maintain robust food safety documentation, and work closely with delivery platforms to challenge suspicious refund claims more effectively.
It comes as AI-generated content becomes increasingly widespread online, with more than 34 million AI-generated images now reportedly produced every day and 71% of social media content estimated to be AI-generated.
Food Alert said the issue of ‘AI food fraud’ is intensified by how many third-party delivery platforms handle complaints, with refunds often processed automatically before costs are passed directly onto restaurants and takeaways.
“There have been instances where we’ve seen AI manipulation of images relating to foreign bodies or undercooked food complaints,” said Food Alert AI lead Alasdair Dean.
“We do think this is likely to increase, and we are aware that third-party aggregators reportedly receive a lot of suspicious complaints of this nature.”
Food Alert also warned that identifying fake complaints is becoming increasingly difficult as AI technology advances.
“At the moment, email and written complaints are relatively easy to identify as AI language is currently fairly easy to distinguish. However, this is likely to change over time. Images are harder to spot, as while there are clues to look for, they are trickier to see. Again, these will become harder and harder to identify over time, and relatively rapidly,” Dean added.
No comments yet