On 22 September 2026, a federal judge in Portland, Oregon, sentenced Karen Vardanyan, a 35-year-old Armenian national, to 24 months in federal prison and three years of supervised release, with $1,219,106 in restitution to the victims, for his role in the Ryuk ransomware conspiracy that struck companies, schools, municipalities and hospitals around the world between March 2019 and mid-2020. Vardanyan had been arrested in Kyiv in April 2025 at the request of the United States, extradited from Ukraine in June 2025, and pleaded guilty on 8 July 2026 to conspiracy and fraud in connection with computers. The indictment had alleged that he and his co-conspirators deployed Ryuk on hundreds of compromised servers, collected roughly 1,610 bitcoins valued at more than $15 million at the time of payment, and launched more than 2,400 attacks in total. The sentence that followed — two years, not the five years per count the statutes allowed — is the first hard data point in one of the most instructive ransomware prosecutions of the decade. But the file is not closed. A second Armenian national, 45-year-old Levon Georgiyovych Avetisyan, charged with the same conspiracy, was arrested in France, where he is the subject of an American extradition request. Two more alleged members, Ukrainian nationals Oleg Nikolayevich Lyulyava and Andrii Leonydovich Prykhodchenko, are charged and not in custody. For anyone arrested in France on a foreign cybercrime request — or advising someone who is — the Ryuk sequence raises the only questions that matter: how does extradition to the United States actually work once you are in a Paris cell, which defences genuinely move a French extradition file, how heavy is the American sentence at the end of the road, and what would French criminal law itself do with ransomware conduct built on unauthorised access, encrypted servers and bitcoin payments?
I. One conspiracy, two fates: jailed in Oregon, wanted in Paris
A. From a Kyiv arrest to 24 months in Portland: how the first Ryuk defendant got there
The public record of the Vardanyan prosecution, set out in two press releases of the United States Attorney’s Office for the District of Oregon, reads like a manual of modern international cybercrime enforcement. According to the first release of 16 July 2025, Vardanyan was placed on an international wanted list at the request of the FBI, arrested in Kyiv in April 2025, extradited from Ukraine to the United States on 18 June 2025, made his first appearance before a federal magistrate judge on 20 June 2025, pleaded not guilty, was ordered detained, and was initially set for a seven-day jury trial. Thirteen months later the posture had changed completely: on 8 July 2026 he pleaded guilty to conspiracy and fraud in connection with computers, and on 22 September 2026 he received 24 months in federal prison, three years of supervised release and a restitution order of $1,219,106. The superseding indictment had been returned by a Portland federal grand jury on 22 February 2024 and charged conspiracy, fraud in connection with computers and extortion in connection with computers, each count carrying up to five years in prison, three years of supervised release and a $250,000 fine. The gap between the exposure — fifteen years in theory — and the 24 months imposed is the whole lesson of the case for defence lawyers: in the American federal system, an early acceptance of responsibility, a negotiated plea and a focused sentencing presentation can compress the outcome dramatically, even in a file the Department of Justice describes as one of the most damaging ransomware waves ever prosecuted.
The scale alleged in the file explains why the Americans pursued it across three countries. Court documents cited by the Department of Justice describe Ryuk as malicious software that encrypts the data on a victim’s computers and servers and blocks all access until a ransom is paid, usually in bitcoin, with a ransom note left on the compromised systems giving the victims an email address to open negotiations and obtain decryption keys. Between March 2019 and September 2020, Vardanyan and his co-conspirators allegedly broke into the networks of victim companies, deployed Ryuk on hundreds of servers and workstations, and paralysed businesses, local authorities, schools, critical infrastructure operators and health-care providers. The indictment alleged receipts of about 1,610 bitcoins, valued at more than $15 million when paid; the specialist press, drawing on the same court documents, reports more than 2,400 attacks worldwide and ransoms that severely disrupted the victims’ ability to function. Ryuk itself was first detected in August 2018, has been linked by investigators to Russian-speaking cybercrime networks connected with the TrickBot malware, and drew a joint warning from American federal agencies in 2020 when its operators turned their attacks on hospitals in the middle of the pandemic. A Russian facilitator accused of laundering Ryuk proceeds had already pleaded guilty in 2023 in the same Oregon court. Against that background, the 24-month sentence looks like the product of an individualised assessment — a defendant convicted as a member of the conspiracy, not as its organiser, who ultimately accepted the facts — rather than a tariff for ransomware as such. Anyone comparing their own exposure with this number should understand that: the American sentence punishes the person’s proven role, criminal history, acceptance of responsibility and ability to pay restitution, not the headline totals of the whole network.
For a French reader the second half of the story matters more than the first. While Vardanyan was flying toward Portland, his alleged co-conspirator Avetisyan was arrested in France, and the United States issued an extradition request to the French authorities for the same charges of conspiracy, computer fraud and computer extortion. That request is now, or should now be, moving through the French passive-extradition procedure in Paris: presentation to the general prosecutor, detention under an extradition warrant, a hearing before the investigating chamber of the Paris Court of Appeal, a reasoned opinion, and then a government decree that alone authorises surrender. Avetisyan is presumed innocent, and everything that follows in this article is written with that presumption in mind: an extradition request is an accusation by a foreign state, not a conviction, and the French procedure exists precisely to test it before any surrender. But the file also shows how fast the American side moves once an indictment exists — FBI wanted notice, provisional arrest abroad, formal request, and, as the Vardanyan track proves, a trial date set within months of arrival. A person arrested in France on an American request who waits to see what happens has already lost the most valuable weeks of the defence. The rest of this article explains what a Paris lawyer does with those weeks.
B. Can the United States really have you surrendered from France for hacking done abroad?
Yes — if the statutory conditions are met, and the Ryuk facts meet most of them on paper, which is exactly why the defence must attack the file point by point rather than hope the request collapses on its own. French passive extradition is governed by the Code of Criminal Procedure, and the first filter is the list of offences that can support surrender. Article 696-3 of the Code of Criminal Procedure provides that extraditable acts are, first, all acts punished with criminal penalties under the requesting state’s law and, second, acts punished with correctional penalties where the maximum prison term under that law is at least two years — or, for a convicted person, where the sentence imposed is at least two months. The American Ryuk charges each carry up to five years in federal prison, so the American side of the threshold is comfortably cleared. But the same article contains the rule that decides most cybercrime files: “En aucun cas l’extradition n’est accordée par le gouvernement français si le fait n’est pas puni par la loi française d’une peine criminelle ou correctionnelle.” No double criminality, no surrender. Ransomware conduct — breaking into servers, encrypting data, extorting bitcoin — is punishable under French law by several years of imprisonment, as Part II shows, so the double-criminality test is unlikely to defeat the request by itself. It still has to be verified offence by offence, because American conspiracy and computer-fraud counts do not always map onto French qualifications, and every count that fails the mapping must be carved out of the surrender through the specialty rule discussed below.
The second filter is the catalogue of mandatory refusals, and it is wider than most requested persons imagine. Article 696-4 of the Code of Criminal Procedure opens with the blunt formula “L’extradition n’est pas accordée” and then lists eight bars: French nationality of the requested person assessed at the time of the offence; political character of the offence or a political purpose behind the request; offences committed on French territory; offences already finally tried in France; time-bars under either law; a penalty in the requesting state contrary to French public policy; trial by a court that does not guarantee fundamental procedural rights and defence rights; and purely military offences. In a Franco-American cybercrime file, three of these bars do real work. The nationality bar protects French nationals but not Armenian or Ukrainian nationals arrested in France. The public-policy bar covers penalties unknown to or rejected by the French legal order — American federal sentences are served in a system the French courts examine concretely, and arguments about irreducible life terms or inhuman detention conditions, which have blocked surrenders to the United States in other European cases, belong here; readers facing an American request built around crushing statutory exposure should study when an American sentence can block surrender from France alongside this article. The fair-trial bar allows the defence to attack the procedure awaiting the person across the Atlantic. None of these arguments wins by incantation; each must be documented, translated, expert-supported and filed in time. That is the difference between a lawyer who knows the extradition chamber and one who discovers it.
The procedure itself is the third thing to understand, because its timetable is the defence’s weapon. Once arrested, the requested person is brought before the general prosecutor, and Article 696-15 of the Code of Criminal Procedure provides that where the person tells the prosecutor he does not consent to surrender, the investigating chamber is seized without delay and he appears before it within ten working days. The decisive sentence follows: “Si, lors de sa comparution, la personne réclamée déclare ne pas consentir à être extradée, la chambre de l’instruction donne son avis motivé sur la demande d’extradition.” The chamber gives a reasoned opinion within a month of the appearance unless further inquiries are ordered, and “Cet avis est défavorable si la cour estime que les conditions légales ne sont pas remplies ou qu’il y a une erreur évidente.” An unfavourable opinion binds the government: there can be no surrender decree against it. A favourable opinion, conversely, does not surrender anyone by itself — only a government decree does, and that decree can be challenged before the administrative courts. The practical consequences are straightforward. First, never consent to simplified surrender without advice; consent waives the chamber’s review. Second, the defence memorandum for the chamber is the most important document in the file, because it frames every later challenge. Third, detention during the procedure is not automatic or endless: Article 696-19 of the Code of Criminal Procedure states that “La mise en liberté peut être demandée à tout moment à la chambre de l’instruction selon les formes prévues aux articles 148-6 et 148-7”, with a ruling within twenty days, or fifteen where the request comes within forty-eight hours of the extradition detention order, and the chamber can attach release to judicial supervision or house arrest with electronic monitoring. The mechanics of staying out of custody while the request is examined are explained in detail in how release pending an extradition ruling works in Paris. Every week spent free is a week to build the file that may stop the surrender.
II. Beating the request in Paris, and what French courts do with ransomware
A. Four defences that actually move a Paris extradition file in a ransomware case
The first defence is double criminality argued count by count, not as a slogan. The investigating chamber must satisfy itself that each American count corresponds to conduct punishable under French criminal law, and American federal cybercrime counts are broader in some respects and narrower in others than their French equivalents. Conspiracy under American law criminalises the agreement itself plus an overt act; French criminal association and conspiracy provisions have their own thresholds of a formed group, a proven agreement and material preparatory acts. Computer fraud and computer extortion under American law group together conduct that French law splits between unauthorised access, data manipulation, extortion and money laundering, each with its own elements. Where a count has no French mirror, surrender for that count must be refused, and the refusal matters even if other counts survive, because of the specialty principle: the requesting state may only try the surrendered person for the offences for which surrender was granted. The French Court of Cassation polices this boundary strictly. In a ruling of 28 May 2024, No. 24-81.539, the criminal chamber recalled that trial judges “sont tenus de s’assurer que la personne placée en détention provisoire ne l’a pas été en violation du principe de spécialité, lorsque l’intéressé n’a pas renoncé à ce principe”, quashed the Paris investigating chamber’s detention order in its entirety, ordered the prisoner’s release and declared him detained without valid title. The case concerned a European arrest warrant chain, but the specialty principle it enforces applies with equal force to classic extradition: “CASSE et ANNULE, en toutes ses dispositions, l’arrêt susvisé” and “ORDONNE la mise en liberté” are the words every extradition lawyer wants to read in his own file. In practice, the Paris defence therefore demands the American indictment, the statutes and the sentencing guidelines in certified translation, maps each count onto French qualifications, and asks the chamber to exclude whatever does not map — narrowing the surrender, and sometimes gutting the requesting state’s case, even where surrender itself cannot be avoided.
The second defence is the concrete review of what awaits the person if surrendered: the sentence, the prison conditions and the fairness of the trial. French courts do not take the requesting state’s assurances at face value; they examine them. The freshest illustration is a Court of Cassation ruling of 27 May 2026, No. 25-84.904, in an extradition requested by Algeria for the enforcement of a life sentence imposed in absentia for organised drug trafficking and money laundering, where the defence argued a risk of inhuman or degrading treatment from collective detention conditions. The criminal chamber rejected the appeal, but the reasoning is the point: it validated the investigating chamber because it had “s’est livrée à un examen concret des conditions de détention fondé sur les éléments fournis par les autorités requérantes à la suite des compléments d’information qu’elle avait ordonnés” and had thus, with sufficient and consistent reasoning, ruled out the risk of undignified conditions — “a souverainement écarté, par des motifs exempts d’insuffisance et de contradiction, l’existence du risque de conditions indignes de détention”. The ruling ended with “REJETTE le pourvoi”. For the defence, the message is double-edged and useful. First, the chamber can and does order further inquiries and demand concrete evidence from the requesting state — square metres per prisoner, hours of outdoor exercise, health facilities — instead of ruling on abstract claims; the defence should therefore force those inquiries by filing detailed, sourced detention-condition evidence about the specific American facilities at issue. Second, the Court of Cassation controls the sufficiency of reasoning under Article 3 of the European Convention on Human Rights, so a chamber that waves the request through with generic formulas commits a reversible error. In an American file, this defence is built from Bureau of Prisons data, sentencing-position papers showing the real exposure after trial versus plea, expert opinions on supermax or special administrative measures where relevant, and any element — health, age, family, length of potential sentence — that makes the concrete situation, not the abstract system, the question. It rarely produces a refusal on its own in American cases, but it regularly produces delays, further inquiries, conditional opinions and negotiating leverage — and leverage is what gets counts dropped and pleas improved.
The third and fourth defences are procedural and factual: mistaken identity or evident error, and liberty. “Une erreur évidente” in Article 696-15 covers the file where the wrong person is wanted, where the facts in the warrant contradict the documents, where the offence is time-barred under French or American law, or where the requesting state’s papers do not establish even a prima facie case linking the arrested person to the counts. Ransomware attribution is notoriously fragile — shared aliases, reused infrastructure, nicknames like the ones cited in the Ryuk file, cryptocurrency trails that pass through mixers and intermediaries — and the defence should hire its own technical expertise early rather than concede the attribution narrative in the American papers. As for liberty, Article 696-19 release applications should be filed immediately and renewed whenever the file evolves: a person who appears free, employed, housed and supervised before the chamber is a person the chamber can imagine leaving free under supervision rather than surrendering from a cell. The chamber may order release under judicial supervision or house arrest with electronic monitoring, and each release order also signals to the government, which decides last, that the file is contested and contestable. None of this requires believing the American case is weak. It requires treating the Paris phase as a trial in its own right — with evidence, experts, translations and witnesses — instead of a waiting room before a flight to the United States.
B. What French law itself calls ransomware: hacking, data crimes, malware tools and conspiracy
American prosecutors charge Ryuk-style conduct as conspiracy, computer fraud and computer extortion. French law reaches the same conduct through the chapter of the Criminal Code devoted to attacks on automated data-processing systems, and the mapping is precise enough that double criminality will usually be satisfied — which is why the defence fights on the margins of the mapping rather than on its existence. The entry point is Article 323-1 of the Criminal Code: “Le fait d’accéder ou de se maintenir, frauduleusement, dans tout ou partie d’un système de traitement automatisé de données est puni de trois ans d’emprisonnement et de 100 000 € d’amende.” Where the intrusion causes deletion or modification of data or alters the functioning of the system, the penalty rises to five years and 150,000 euros — the exact factual pattern of a ransomware deployment, which by design alters the system and renders data inaccessible. Deploying the payload itself falls under Article 323-3 of the Criminal Code, which punishes with five years and 150,000 euros “Le fait d’introduire frauduleusement des données dans un système de traitement automatisé, d’extraire, de détenir, de reproduire, de transmettre, de supprimer ou de modifier frauduleusement les données qu’il contient”. Encrypting a victim’s files with Ryuk is fraudulent introduction and modification of data within the meaning of that text; exfiltrating copies to a leak server is fraudulent extraction and transmission. Supplying or sharing the ransomware builder, the encryptor, the decryption infrastructure or the negotiation panel is caught by Article 323-3-1 of the Criminal Code: “Le fait, sans motif légitime, notamment de recherche ou de sécurité informatique, d’importer, de détenir, d’offrir, de céder ou de mettre à disposition un équipement, un instrument, un programme informatique ou toute donnée conçus ou spécialement adaptés pour commettre une ou plusieurs des infractions prévues par les articles 323-1 à 323-3 est puni des peines prévues respectivement pour l’infraction elle-même ou pour l’infraction la plus sévèrement réprimée.” And the collective dimension — developers, intruders, negotiators, launderers — is covered by Article 323-4 of the Criminal Code: “La participation à un groupement formé ou à une entente établie en vue de la préparation, caractérisée par un ou plusieurs faits matériels, d’une ou de plusieurs des infractions prévues par les articles 323-1 à 323-3-1 est punie des peines prévues pour l’infraction elle-même ou pour l’infraction la plus sévèrement réprimée.” Added to this core are extortion, blackmail and money-laundering provisions for the ransom demands and the bitcoin trail. The comparison with the parallel Franco-British ransomware extradition file examined in the recent KillSec extradition case and its French defence lessons confirms the pattern: European investigating chambers have no difficulty finding French mirrors for foreign ransomware counts, so the battle moves to specialty, sentencing exposure and detention conditions.
Two practical consequences follow for anyone arrested in France in a file like Ryuk. First, the French qualification also determines what happens if extradition is refused: France can prosecute on its own territory conduct committed abroad by a person present in France under certain conditions, and a refusal of surrender does not always mean the end of criminal exposure — it can mean a French prosecution instead, before judges applying the very articles quoted above. A defence strategy built entirely on blocking surrender without preparing the fallback of a French case is incomplete. Second, the evidence gathered abroad — seized servers, terabytes of exfiltrated data, cryptocurrency tracing, chat logs from seized infrastructure — will be challenged under French evidentiary rules if a French trial follows, and under American suppression rules if surrender follows: chain of custody across three or four countries, legality of the foreign interceptions, reliability of blockchain attribution, and the conditions in which any statements were obtained. The Ryuk file shows investigators pooling evidence from Ukraine, the United States and Europol-supported operations; each transfer is a link the defence can test. Preserve everything now: devices, accounts, correspondence with alleged co-conspirators, proof of legitimate employment and income, travel records contradicting attribution windows, and the names of witnesses who can explain what the technical traces really mean. By the time the investigating chamber holds its hearing, it is too late to start collecting.
Conclusion
The Ryuk sequence delivers three lessons that will outlast the headlines. First, international ransomware enforcement now runs end to end: FBI wanted notice, arrest in a third country, extradition, guilty plea and federal sentence within about seventeen months, with the 22 September 2026 Portland sentence — 24 months, three years of supervised release and $1,219,106 in restitution — showing that even the flagship cases end in individualised, negotiable outcomes rather than the stacked statutory maximums of the indictment. Second, the French phase of such a file is a genuine trial before surrender, not a formality: double criminality count by count, the specialty shield enforced by the Court of Cassation, concrete review of detention conditions and trial fairness, evident-error challenges to fragile attribution, and release applications from the first forty-eight hours. The Court of Cassation’s 27 May 2026 ruling demanding concrete examination of prison conditions and its 2024 specialty ruling quashing a detention and freeing the prisoner prove that Paris judges decide these files on evidence, not on deference. Third, none of this helps a person who waits: refuse nothing and concede nothing without counsel, demand translations of every American document, commission independent technical and prison-conditions expertise, and file for release immediately. Levon Avetisyan, like every requested person, is presumed innocent, and only the courts — French courts first, on the surrender, and then, if surrender is ordered, the American court on the merits — will decide his fate. What is certain is that the weeks after the arrest in France decide more than the months of procedure that follow. They should be spent building the file, not watching it move.