Foodora Denies Refund Over Suspected AI-Generated Photo

Food delivery platforms and e-commerce firms are increasingly using AI-detection claims to deny customer refunds, as seen in a recent dispute between Foodora and a McDonald’s customer reported by Tønsbergs Blad. This trend highlights a growing tension where companies allege AI-generated fraud to avoid payouts for damaged goods.

A customer who ordered McDonald’s via Foodora was denied a refund after the company claimed photos of the damaged meal were AI-generated. According to a report by Tønsbergs Blad published June 17, 2026, the customer complained that a drink had spilled, turning their Big Mac into what they called a “Wet Mac.” Foodora refused the reimbursement, citing suspicions that the evidence was fabricated using artificial intelligence.

How is AI changing the refund process for consumers?

Companies are shifting from a “trust-but-verify” model to a “verify-then-trust” approach. In the Foodora case reported by Tønsbergs Blad, the burden of proof shifted from the company proving the food was fine to the customer proving the photo wasn’t a deepfake. This creates a new friction point in the gig economy where digital evidence is the primary currency for dispute resolution.

This shift mirrors trends in the insurance industry. According to reports on insurance fraud trends, carriers are now using metadata analysis and AI-detection software to spot “synthetic” accident photos. When a company like Foodora claims a photo is AI-generated, they aren’t just questioning the meal; they’re questioning the authenticity of the digital file itself.

Why are companies accusing customers of using AI?

The primary driver is the accessibility of high-quality image generation tools. It now takes seconds to create a realistic image of a spilled soda or a crushed package. For platforms managing millions of deliveries, these “micro-frauds” add up to significant losses. By flagging suspicious images as AI-generated, companies can reduce their refund ratios and lower operational costs.

However, this creates a “false positive” risk. If a company’s AI-detection tool flags a genuine photo of a soggy burger as “synthetic,” the customer has little recourse. The Foodora incident shows that a simple photo—once the gold standard of proof—may no longer be sufficient to secure a refund.

What happens when “AI-generated” becomes a corporate shield?

There is a growing concern that “AI-generated” will become a catch-all excuse for companies to deny legitimate claims. When a company claims a photo is fake, the customer is rarely given the technical data to prove otherwise. This asymmetry of information puts the consumer at a disadvantage.

Legal precedents regarding digital evidence are still catching up. While traditional consumer protection laws require companies to provide services as described, the introduction of “synthetic media” complicates the definition of evidence. If a platform denies a refund based on an AI-detection algorithm, the question becomes: who audits the algorithm?

Comparison: Traditional Fraud vs. AI Fraud

FAQ: AI and Consumer Refunds

Can companies legally deny a refund if they suspect AI?
Generally, companies can deny claims based on their terms of service regarding fraud. However, consumer protection agencies may intervene if the company cannot provide evidence of the fraud.

How can I prove my photo is real?
Keep the original file with its EXIF data (metadata that shows the time, date, and device used). Providing a video or a photo of the item next to a current newspaper or a digital clock can also help.

Is AI-detection software 100% accurate?
No. AI detectors often produce false positives, especially with low-resolution photos or images with unusual lighting, which can lead to legitimate claims being rejected.

For more on the intersection of technology and consumer rights, see our guide on digital evidence laws or visit the Federal Trade Commission for tips on handling corporate disputes.