Report finds 2,100 AI recipe sites, including one using an Emily Blunt lookalike

Report finds 2,100 AI recipe sites, including one using an Emily Blunt lookalike

A report flags 2,100 suspected AI recipe sites posing as real cooks’ blogs, including one using an AI-generated Emily Blunt lookalike.

A baking blog uses an AI-made likeness of Emily Blunt as its supposed author, according to Futurism’s coverage of a Decrypt Ads investigation. The report identifies more than two thousand sites, including this one, that strongly appear to be AI slop farms pretending to be blogs run by actual chefs and home cooks.

A startup that follows online advertising supply chains, Decrypt Ads published its report on the day Futurism covered the findings. The investigation found fake recipes from these sites appearing in Google searches and on social platforms such as Pinterest. The sites frequently rely on sketchy Russian and Chinese adtech partners to make money, according to the report.

According to the report, it’s “common knowledge that publishers in certain niches are being openly deceptive about their owners and authors, using AI photos and fake back stories on their sites.” It adds: “Some of the most consistent growth in AI slop spreading across the internet has come from AI recipe websites.”

Manual searches provided the starting point for the investigation. Decrypt Ads researcher and security expert Zach Edwards told Futurism that his initial search uncovered about 400 questionable sites publishing what looked like AI-generated recipes under fictitious chefs’ names. Using “technical fingerprints,” he then found hundreds of additional sites associated with the suspicious domains he had manually uncovered. The resulting count reached roughly 2,100 separate domains, with many seemingly controlled by the same operators.

Edwards said: “We really approach this expecting these operators to try and hide their websites.” He continued: “These folks they don’t want people to know all of the websites they own.”

The sites also work to attract an audience through social media. Their strategies are often aggressive, the report finds. Edwards called the fake brands “super aggressive on Pinterest, super aggressive on Twitter, on Facebook.”

He added: “Any place where there’s visual media,” followed by, “you can expect these AI author people to be trying to drum up organic traffic.”

The report counts AddCakes among its most aggressive examples. The baking site claims it’s operated by a baker named Elena, describing her as a “busy mom who wanted to connect with my roots through baking, but got overwhelmed by complicated cake recipes and the pressure to make them look perfect.”

Yet Elena is plainly an AI-created likeness of Blunt, a highly recognizable British actress whose films include “The Devil Wears Prada” from 2006 and “A Quiet Place” from 2018. Futurism describes this appropriation of her appearance as a brazen act of theft.

A profile photo isn’t the only part of the invented persona. Images published by AddCakes show Elena preparing food in a kitchen and seated for dinner alongside a family generated by AI. Futurism offers two possible explanations: the operator expects readers not to spot the altered celebrity pictures, or hopes Blunt’s face will bring them to the site.

According to location data from X, previously known as Twitter, the unidentified person behind the operation is based in Morocco. That’s not where Blunt lives. Futurism sought comment from her representatives but received no response.

Edwards said websites working with Russian and Chinese adtech partners carry a privacy risk for consumers because authoritarian regimes may seek to collect valuable browsing data.

As he put it: “This report is about publishers who are being deceptive. Their audiences could be impacted, advertisers are wasting money.” He added: “But the underlying fact is also that if you load this website and you don’t block ads, your data is going to, basically, authoritarian countries.”

The problem, in Futurism’s assessment, isn’t limited to a web clogged with spam. These sites combine potential security risks with dishonest assertions of expertise. Their fictional biographies suggest trained cooks created and tested the recipes, despite offering no basis to establish that anybody has prepared them.

Edwards explained: “Most of these AI recipe websites, when you look at those ‘about’ pages, the fake backstories that they’re putting together… this lady was a Michelin star chef and she worked at this fancy hotel or this place, and they’re basically saying that this expert not only wrote these recipes, but tested themselves. And it’s giving this air of, ‘Wow, you’re about to get a recipe from an expert. This has all been carefully vetted,’” He added: “But none of these recipes have probably been cooked – zero of them by the actual owners or authors, and to me that’s the scary part.”

Many of these sites are likely scraping content ideas from real bloggers and creators who have taken the time, energy and expertise to test, photograph and publish real recipes, Edwards said.

AddCakes supplies Elena with a backstory, kitchen photos and even an AI-generated family, but none of that establishes that anyone has cooked its recipes. The site builds a detailed picture of the baker while leaving the baking unverified.

The Genius Geek
The Genius Geek

The Weekly Geek

Don't miss a thing: get our top stories in a weekly roundup.