A review of Telegram communities involved in explicit nonconsensual content finds 50+ bots that claim to create such content, listing 4M+ total “monthly users”
We identified more than 50 Telegram bots built to create nude images and video. After sending questions to Telegram about their harm, the company removed the bots …
Context & Ripple Effects
This is a recurrence, not a newly discovered abuse category: 2020 reporting documented a Telegram bot that turned photos into fake nudes and found more than 100,000 altered images shared in public channels. That earlier discovery established both the low-friction distribution model and the consent harm at issue.
The fact that a similar service was reported as still operating a month later despite its initial exposure makes this review consequential: it tests whether platform removals can keep pace with bot-based creation services rather than merely remove identified accounts.
First-order effects
- Telegram removed the more than 50 identified bots after researchers’ inquiries, cutting off access to those specific services within the app.
- Users of those bots and people targeted by their outputs face an immediate disruption in creation and distribution, though the reported user total is self-reported by the bots and not an independently verified audience measure.
Second-order effects
- The removals increase the burden on Telegram to detect successor bots and closely related channels proactively; a complaint-led response leaves enforcement dependent on outside researchers identifying services.
- Other messaging and bot platforms may face greater scrutiny over whether their automation tools and discovery features enable nonconsensual-image services, an issue tied to the earlier fake-nude bot reporting.
Third-order effects
- If repeated removals remain reactive, explicit-deepfake abuse is likely to become a continuing platform-governance problem centered on account creation, bot discovery, and repeat-offender detection rather than a one-time content takedown issue.
- The pattern strengthens pressure for clearer access and accountability rules around tools that transform a person’s image without permission, though the corpus does not establish which regulatory approach, if any, will prevail.
The trend: This is one data point in the shift from isolated deepfake tools toward scalable, bot-mediated nonconsensual-image services that force platforms to treat access controls as a core safety function.