Ever had a company promise you a gourmet meal and then serve you a single, sad cracker? That’s the vibe right now with Perplexity AI, where asking for the menu might just get you kicked out of the restaurant.

A user on their official subreddit noticed the “Deep Research” feature, a big selling point for the Pro subscription, wasn’t digging very deep. In a move of pure internet detective work, they used Perplexity’s own official documentation and launch blog to prove the tool was severely throttled. The post blew up, with hundreds of upvotes and comments from fellow users agreeing. Then, instead of a “thanks for the feedback,” the response was a permanent ban and a deleted thread. Poof. Problem silenced.

It’s the classic tale of shooting the messenger, but the messenger brought receipts stamped by the company itself. There’s something deeply funny about using a company’s words to call them out, only for them to respond by covering their ears and yelling “la la la, can’t hear you!” It’s like getting banned from a library for reading a book too accurately.

The real kicker is the moderation strategy. Nothing says “we stand by our product” quite like erasing the evidence and banning the person who found it. It’s the digital equivalent of “if I can’t see you, you don’t exist.” They turned a customer service moment into a viral cautionary tale, which is arguably a much harder piece of marketing to pull off.

In the end, this is a perfect snapshot of modern internet culture: a blend of consumer advocacy, hilarious overreaction, and the timeless lesson that you should never bring a company’s own receipts to a fistfight. Just be ready for them to change the locks.