Two Legal Theories Behind Pavel Durov's Terrorism Charges

Last-verified risk record: two criminal theories, not one

Any legal analysis of the charges against Pavel Durov has to start by separating two live criminal tracks that are easy to collapse into one headline. As of Q3 2026, France is testing an untried cybercrime complicity theory against Telegram’s CEO. Russia is using a terrorism-facilitation provision after a longer campaign against Telegram’s independence from state control. Neither track is a conviction record. Neither should be treated as interchangeable with the other. The practical risk question is narrower: when does a platform’s failure to cooperate with law-enforcement demands become a prosecutor’s theory of personal criminal liability for the named executive who controls the service?

France’s LOPMI theory turns platform administration into the hinge

The French case matters less because it involves Telegram, and more because of the statutory move it attempts. Reuters described the LOPMI provision as a tough, untested cybercrime law with no direct Western analog identified by former U.S. deputy assistant attorney general Adam Hickey in the reporting [2]. That is a materially different starting point from the familiar intermediary-liability debate over whether a platform should be treated as the speaker, publisher, distributor, or passive host of user content.

The LOPMI allegation, as reported, is aimed at “complicity in the administration of an online platform to allow an illicit transaction, in an organized gang” [2]. The word doing the work is administration. A prosecutor does not have to say that the CEO wrote the criminal messages, sold the contraband, or personally recruited the users. The theory instead looks to the control layer: who operated the platform, who could change the cooperation posture, who received or ignored official requests, and whether the administration of that system allowed illicit transactions to continue.

That is why the unanswered-request record is not a side fact. Reuters reported that the Paris J3 cybercrime unit, led by Johanna Brousse, built the case through an undercover operation and that France’s National Gendarmerie counted 2,460 unanswered law-enforcement requests over 11 years [2]. The number does not itself prove criminal complicity. It does, however, give prosecutors a way to argue that non-cooperation was not an occasional breakdown, a technical misrouting problem, or a misunderstanding over one request. It becomes the evidentiary spine for an administrative-facilitation theory.

The chronology is short enough to state without turning it into biography. Durov was arrested in France on Aug. 24, 2024, and indicted on Aug. 28, 2024, in a 12-count case that included allegations tied to illegal transactions, child sexual abuse material, narcotics, fraud, and non-cooperation with authorities [1]. Bail was set at €5 million, and travel restrictions followed [1]. France 24 later reported that France fully lifted his travel ban on Nov. 13, 2025, while the case remained under investigation [6].

The important distinction is procedural. An indictment is not a finding that Telegram’s architecture caused the charged offenses, or that Durov personally intended each downstream crime. The French case is still an active investigation in the materials provided, with no trial date and no conviction. The risk signal is that prosecutors are treating the platform’s administrative refusal pattern as potentially criminal conduct by the person at the top of the chain, rather than only as a basis for fines, takedown orders, or civil regulatory sanctions.

The Coco comparison belongs here, but only as context. Reuters reported that the same LOPMI framework was applied to Isaac Steidel, founder of the Coco chat forum, which was linked to more than 23,000 legal proceedings [2]. That does not make Telegram equivalent to Coco. It does show that French authorities were not treating Article 323-3-2 as a one-off Telegram instrument. The provision is being positioned as a way to reach platform administrators where investigators say user crime is organized through the service and official requests do not produce cooperation.

Durov’s own public posture fits the legal dispute rather than resolving it. France 24 reported that he called the French charges “misguided,” argued that prosecutors could have used more targeted requests, and that Telegram later increased moderation and launched AI moderation tools after the August 2024 arrest [6]. Those facts can support different arguments. The defense can point to overbroad state demands and post-arrest remediation. Prosecutors can point to the same remediation as evidence that cooperation and moderation choices were available all along. Neither point decides the statutory question.

Russia’s Article 205.1.1 case requires a different evidentiary register

Russia’s case is not simply the French theory translated into harsher vocabulary. Euronews reported that Russia opened a criminal case against Durov in February 2026 for promoting terrorism [3]. By July 29, 2026, Russian authorities had issued an international arrest warrant under Article 205.1.1, according to TechTimes reporting, with the case framed around refused content takedowns and terrorist use of Telegram [4]. AP separately reported that Russia accused the Telegram CEO of aiding terrorism and corrected the potential sentence to life imprisonment [5].

The factual allegations are severe, but their source matters. TechTimes and AP reported FSB claims that more than 153,000 crimes had been committed through Telegram since 2022, including more than 33,000 sabotage, terrorism, or extremism-related offenses [4][5]. The same cluster of reporting attributes to Russian authorities claims involving the Crocus City Hall attack, in which 145 people were killed, and the killings of Darya Dugina, Vladlen Tatarsky, and Igor Kirillov [4][5]. In a risk record, those are attributed allegations by a state security service, not adjudicated facts about Durov’s criminal responsibility.

The chatbot allegation should be handled the same way. TechTimes reported Russian claims that a dating chatbot identified as Dayvinchik or Leo recruited Russians aged 12 to 22 for sabotage, and that 46 people had been detained since July 2025 in connection with that alleged recruitment channel [4]. If credited, the allegation helps explain why Russian authorities chose terrorism-facilitation language. It should still be read as a state allegation in an ongoing case, not as an independent finding that Telegram’s CEO facilitated those acts.

The surrounding enforcement environment is also different. TechTimes reported that Telegram had been degraded to roughly 95% blockage in Russia and that the MAX messenger had been mandated as part of the domestic alternative environment [4]. Those facts matter because they place the warrant inside a broader digital-sovereignty campaign. They do not make every Russian allegation false. They do make the procedural safeguards, source independence, and institutional incentives materially different from the French record.

That is the point at which terrorism language becomes legally dangerous to read too quickly. Terrorism facilitation is not just a harsher label for bad moderation. It can alter bail, sentencing exposure, diplomatic posture, and the public tolerance for procedural shortcuts. In Russia’s presentation, Telegram is not merely a platform that failed to answer requests; it is cast as infrastructure used by enemies of the state. For lawyers tracking executive exposure, that is a different category of prosecutorial leverage.

The convergence is real, but the systems are not equivalent

The two cases should not be flattened into a single proposition that governments everywhere are now criminalizing encrypted messaging. France’s theory is a first-impression Western prosecution under a cybercrime provision enacted in January 2023, tested through a criminal investigation and reported legal analysis [2]. Russia’s theory is a terrorism-facilitation case built on claims attributed to the FSB, against the background of blockage, pressure for domestic alternatives, and a July 2026 warrant [4][5]. Those are not the same institutional setting.

They do, however, converge in what they advertise to other prosecutors. Both tracks frame the CEO not merely as the owner of a neutral communication layer, but as a personally responsible node in a system through which user misconduct allegedly continued. France reaches that point through platform administration and ignored requests. Russia reaches it through aiding-terrorism allegations and asserted refusal to take down or cooperate over content. The doctrinal paths are different; the executive-risk destination is close enough to matter.

For platform counsel, the operational lesson is not that every law-enforcement request must be granted, or that every government allegation should be accepted at face value. The lesson is that the request-response record is becoming prosecutorial evidence. Who received the request, what legal basis it cited, how it was escalated, why it was denied or narrowed, and whether the same category of request went unanswered over time may later become the difference between a defensible moderation dispute and an alleged facilitation pattern.

That is especially relevant for legal-tech and compliance teams handling user-generated content, encrypted or semi-encrypted channels, marketplace tools, messaging features, or discovery platforms that can be used by third parties for unlawful conduct. The French record makes the administrative layer visible. The Russian record shows how the same control points can be recast under national-security language when the state chooses a terrorism frame.

Status and enforceability narrow the near-term risk

The immediate enforcement picture is more constrained than the charging rhetoric suggests. The French case remains open in the provided materials, with no conviction and no trial date supplied. France 24 reported that the French travel ban was fully lifted on Nov. 13, 2025 [6]. Russia’s July 29, 2026 warrant carries far more severe stated exposure, but the extradition path is not straightforward: the record provided states that the UAE has no extradition treaty with Russia and that France does not extradite its own nationals [1].

So the near-term risk is not best described as inevitable transfer from one jurisdiction to another. It is the normalization of two different charging templates for platform-executive liability. One is a structured, untested French complicity theory built around administration of a platform and unanswered law-enforcement requests. The other is a Russian terrorism-facilitation theory built around state-security allegations, content takedowns, and digital-sovereignty pressure. Risk counsel now have to track both templates without treating them as legally or institutionally equivalent.

References

1. Arrest and indictment of Pavel Durov — Wikipedia
2. France uses tough, untested cybercrime law to target Telegram's Durov — Reuters, September 17, 2024
3. Russia opens criminal case against Telegram's Durov for promoting terrorism — Euronews, February 24, 2026
4. Russia charges Telegram's Durov with terrorism over refused content takedowns — TechTimes, July 29, 2026
5. Russia accuses Telegram CEO Pavel Durov of aiding terrorism — AP News, July 2026
6. France fully lifts travel ban on Telegram founder Durov — France 24, November 13, 2025