When AI starts pretending to be your PR team, it's a problem
Tech.eu Cate Lawrence
A UK PR firm invented fake staff with AI headshots to pitch journalists. Clients had no clue, and one CEO says most of his inbox looks the same.
Based on reporting by Tech.eu, Cate Lawrence — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
It started the way these things always do lately: half a dozen people forwarding the same link over WhatsApp. The article, from Futurism journalist Maggie Harrison Dupré, detailed how UK-based Movchan Agency had built out more than a dozen fictional PR representatives — complete with names, email addresses, biographies, and in some cases AI-generated or stock-photo headshots — to pitch stories to reporters.
What makes this odd is that Movchan Agency wasn't some fly-by-night operation running purely on fabrications. Plenty of journalists, including this one, had received entirely legitimate pitches from the agency over the years, ones that led to real interviews with real founders. So the question isn't really whether Movchan does good work sometimes — it clearly does — but why it apparently felt the need to invent phantom staffers on top of that. Founder Nadya Movchan told Futurism the fake personas traced back to a domain reputation issue, possibly following a Gmail deliverability update, and that the agency bought pre-warmed domains as a stopgap, which came pre-loaded with fictional names and photos from a vendor. But journalists found pitches from these personas dating back through 2025, and freelancer Ellen Chang said she caught one requesting a meeting in LA with a founder who was actually attending a conference in Vegas — hardly the profile of a short-term technical fix.
The clients, according to Harrison Dupré, had no idea any of this was happening. Movchan Agency says it will wind the practice down, though it's hard not to notice the timing lines up with getting caught rather than a change of heart. Dimitri Masin, CEO of Gradient Labs and a Movchan client, said he was surprised and disappointed when told — but he didn't think this was an isolated case. He estimated that a large share of the unsolicited pitches landing in his inbox now come from fake identities generated this way, and pointed to Artisan's
My take — AI-written commentary, not fact-checked reporting
Fake bylines and AI headshots aren't really an AI story, they're an accountability story: someone decided a synthetic name was easier to manage than an actual human who might get blamed. The unsettling part isn't that one agency got caught, it's that a client-side CEO thinks a majority of his inbox already looks like this, which suggests the industry noticed the trick works before it noticed the trust it burns. Automation is cheap and trust is expensive, and right now a lot of PR and outreach seems happy to spend the expensive thing to save a little of the cheap one.
Read more about this at: Tech.eu
