AI Forensics: Thousands of men in Telegram groups trading nonconsensual images, spyware and doxing

The findings

It has been reported that researchers at the algorithmic auditing nonprofit AI Forensics analyzed nearly 2.8 million messages across 16 Italian and Spanish Telegram communities and found more than 24,000 members posting 82,723 images, videos and audio files over six weeks. The research, summarized in Wired, flags a sprawling underground marketplace where abusive content and services circulate openly — from manipulated intimate photos to paid “spy” tools. These are not fringe chats. They are large, active communities with real victims.

What they're sharing

The content cataloged runs from nonconsensual intimate images and “nudifying” services to folders sellers allegedly claim include child sexual abuse material, incest and rape. It’s grim. And personal. Silvia Semenzin of AI Forensics warns the majority of victims are ordinary women — often people the perpetrators know — who sometimes don’t even know their pictures have been shared or altered. Who gets hurt most? Friends, partners, exes. The emotional punch here: this is privacy invasion that quickly becomes violence.

Hacking, spyware and doxing on offer

Researchers counted more than 18,000 references to spying or spy-related services. Posts promising “professional hacking on commission” allegedly advertise access to phone galleries, anonymous social-media account takeovers and doxing victims by sharing profile links and phone numbers. Buyers ask for bots that can extract private photos. It reads like a how-to for digital stalking. Easy to acquire, easy to weaponize — and, the report suggests, often used against people in the perpetrators’ own circles.

Platform pushback and a regulatory tug-of-war

Telegram says it removes “millions” of items daily using custom AI tools and points to policies banning nonconsensual imagery, doxing and illegal sales. Meanwhile, Pavel Durov is publicly fighting state blocks and quietly defending a platform long accused of harboring extremist and criminal content; he is also under criminal investigation in France for alleged activity on the app. AI Forensics argues Telegram should be classed as a “very large online platform” under EU online-safety rules to force stricter oversight. Free-speech rhetoric meets real-world harm — and the question now is whether regulators will step in before more lives are upended.

Sources: wired.com