Telegram Accuses Ransomers of Using Hidden Pornographic Content to Trigger Temporary Removal from Apple Store
Pavel Durov, founder of Telegram, stated that Apple temporarily removed Telegram from the App Store last night due to users posting illegal pornographic content in public groups, but it was restored within a few hours.
Durov claimed that attackers edited old messages in active groups to insert AI-modified illegal content, making it "invisible" to ordinary members, thereby bypassing manual and automated reviews and triggering the platform's reporting process.
He defined such opponents as "removal ransomers," who use automated accounts to post violations in public groups and report them to Apple, attempting to force groups or communities to pay ransom through the app store review mechanism.
Source: Public Information
ABAB AI Insight
The core of this incident is not merely content review, but rather the "weaponization of platform governance": attackers are no longer directly targeting Telegram, but instead exploiting Apple's App Store review mechanism to turn violations into triggers for removal.
Similar cases can be compared to Telegram's removal by Apple in 2018 due to CSAM content, and subsequent various "report-driven removals"—the difference this time is the addition of AI rewriting, historical message editing, and hidden display, making malicious content harder for ordinary users and automated systems to identify in a timely manner.
In terms of capital pathways, this type of "removal ransom" is essentially a low-cost, high-leverage extortion model: attackers only need to deploy bulk accounts and content to convert the platform's compliance risks into real commercial pressure, forcing community managers or platform operators to pay for stability.
This essentially represents an upgrade of "review attacks" within the context of regulatory changes: shifting from single-point violations to process manipulation, with the underlying mechanism being the linkage between UGC platforms and app stores, making any high-risk content potentially amplified into a distribution-level event by external actors, requiring platforms to expand the "review subjects" from the content itself to the risk of "review processes being abused."
ABAB News · Cognitive Laws
The most dangerous attack is not posting content, but manipulating rules.
Once the review system can be weaponized, reporting will become a business.
In the AI era of content governance, first prevent people, then prevent content.