A viral Reddit post claiming to be from a whistleblower at a major food delivery app has been exposed as an AI-generated hoax. The post, which gained widespread attention with over 87,000 upvotes and was shared across platforms like X, described a fictional insider’s account of systemic exploitation of drivers and users. The author claimed to be a former employee who was writing from a library, using public Wi-Fi, and was allegedly drunk while revealing how the company used AI to manipulate driver incentives and siphon off tips. The story was convincing in part because some of its claims mirrored real events—DoorDash had previously faced a lawsuit over tip theft, ultimately settling for $16.75 million. This connection made the post appear credible, even though the details were fabricated. Casey Newton, journalist behind the newsletter Platformer, reached out to the Reddit user, who responded via Signal. The poster shared a photo of what appeared to be an UberEats employee badge and an 18-page internal document describing a fictional “desperation score” algorithm used to pressure drivers. However, Newton soon realized the post was a carefully constructed scam. He noted that the document’s level of detail would have been time-consuming to create manually—making it unlikely to be a simple troll. But the use of AI tools made it far easier to generate. When Newton used Google’s Gemini with SynthID, a watermarking system, he confirmed the badge image was AI-generated, even after it had been cropped and compressed. Max Spero, founder of Pangram Labs, a company focused on detecting AI-generated content, said the rise of generative AI has made online deception more sophisticated and harder to catch. “AI slop on the internet has gotten a lot worse,” he said, adding that some companies now use AI to create fake engagement, including viral posts that mention their brand, to boost visibility. While tools like Pangram can flag AI-written text, they are less effective with images, videos, and other multimedia. Even when a hoax is debunked, the damage is often done—viral content spreads quickly, and by the time it’s exposed, it has already reached millions. In this case, the incident wasn’t an isolated one. Multiple AI-generated food delivery app hoaxes surfaced on Reddit over the same weekend, each with similar themes and fake evidence. The episode underscores a growing challenge: in an age of advanced AI, verifying online content requires constant vigilance. Readers and journalists alike must now approach viral stories with skepticism, treating every post like a potential forgery until proven otherwise.