KOHEN AVOCATS FIRM · PARIS 16
KillSec Takedown: Eurojust Arrests Teenage Ransomware Chiefs — Would Paris Surrender Them?

Hassan KOHEN acts in urgent matters, from the police station to the assize court. Initial strategic analysis: €80 incl. VAT, personal reply within 24 hours.

100% confidential · Legal professional privilege · No obligation

Article generated by artificial intelligence, through a process designed and supervised by the firm

Source: French Court of Cassation – Judilibre and Légifrance open data

Paris Bar Police custody, investigation, assize court CNB listing avocat.fr
Maître Hassan KOHEN, avocat au Barreau de Paris
Hassan KOHEN
Attorney at the Paris Bar

KillSec Takedown: Eurojust Arrests Teenage Ransomware Chiefs — Would Paris Surrender Them?

Speak to a lawyer 06 89 11 34 45

On 1 October 2026, Eurojust announced the takedown of a ransomware crew that had haunted companies around the world for two years. The group, known as KillSec, is described as “responsible for almost 1 000 attacks worldwide”. Nine countries took part in the operation, three suspects were arrested, eight houses were searched, five servers were seized, and investigators secured “at least 110 terabytes of stolen data”. The most striking detail sits in the middle of the press release: a 16-year-old identified as the group’s “main operator”, while another suspect described as a developer having only just turned 18, after being a minor at the time of several of the alleged offences. For an international audience, the story raises an immediate practical question. KillSec’s members were found in Spain, Greece, the United Kingdom and Romania. Suppose the next operator in a case like this is found in Paris instead. Would France hand him over to the country running the prosecution, or would the suspect stay to face French courts? And what happens when the requesting country is the United States rather than a European neighbour? This article answers those questions from inside French procedure: how a European Arrest Warrant is examined in Paris when the suspect is a teenager, how an American extradition request is filtered by French nationality and territoriality shields, which French criminal charges a ransomware business attracts, and how the seized servers and the cryptocurrency trail are tested once the file reaches a French courtroom. Every suspect named in this piece benefits from the presumption of innocence, and the Eurojust investigation is expressly continuing.

I. Caught in Paris After a KillSec-Style Sweep: Would France Hand the Suspect Over?

French surrender law draws a hard line between two worlds. Inside the European Union, surrender runs through the European Arrest Warrant, a fast judicial channel with narrow refusal grounds. Outside the Union, classic extradition applies, a slower procedure in which the French government keeps the last word and French nationality alone blocks surrender. A KillSec-type file could fall on either side: a German or Spanish judge investigating the servers could send a warrant to Paris, while the United States Attorney’s Office — which took part in the real operation through its Puerto Rico office and the FBI’s San Juan field office — could ask Paris for extradition. The two tracks deserve separate treatment, because the defences available in each are not the same.

A. Would a Paris Court Surrender a Teenage Ransomware Suspect on a European Arrest Warrant?

The starting point is the legal definition. French law provides that “Le mandat d’arrêt européen est une décision judiciaire émise par un Etat membre de l’Union européenne, appelé Etat membre d’émission, en vue de l’arrestation et de la remise par un autre Etat membre, appelé Etat membre d’exécution, d’une personne recherchée pour l’exercice de poursuites pénales ou pour l’exécution d’une peine ou d’une mesure de sûreté privative de liberté.” That definition, found in Article 695-11 of the Code of Criminal Procedure, matters for ransomware because it covers surrender both for trial and for serving a sentence. A German warrant issued by the Hamburg prosecution service against a suspect living in Paris would therefore be examined by the Paris chambre de l’instruction under exactly the same rules as the Italian warrant examined in Rennes in a case decided by the Cour de cassation on 7 August 2024.

That Rennes case, Criminal Chamber, 7 August 2024, appeal no. 24-81.863 (full decision on the Cour de cassation website), is worth reading before predicting any outcome. The file concerned a Romanian national wanted by Italy under a warrant issued on 12 October 2023 for aggravated theft, and the defence plea put to the Court alleged “la violation des articles 695-24, 2°, et 695-33 du code de procédure pénale”. In plain terms, the fight was about Article 695-24, the provision that lets — but does not force — the French court to refuse surrender. The lesson for a KillSec suspect is direct: even inside the streamlined warrant system, the Paris court has a menu of optional refusals, and each one is litigated hard. The judgment shows that questions such as whether the person could serve the sentence in France instead of being surrendered generate supplementary inquiries and full appellate review, not rubber stamps.

Mandatory refusals come first, and one of them speaks directly to the ages in the KillSec file. Article 695-22 of the Code of Criminal Procedure provides that “L’exécution d’un mandat d’arrêt européen est refusée dans les cas suivants”, and its third ground states: “Si la personne recherchée était âgée de moins de treize ans au moment des faits faisant l’objet du mandat d’arrêt européen”. A suspect who was under thirteen at the time of the attacks cannot be surrendered, period. The KillSec main operator is sixteen and the developer passed eighteen, so this shield would not protect them, but it draws the lower boundary every teenage-warrant case must check first. The same article also blocks surrender where a French or another Member State court has already finally judged the same facts, provided any sentence was served or can no longer be enforced, and where the warrant targets someone because of sex, race, religion, ethnic origin, nationality, language, political opinions, sexual orientation or gender identity.

The optional refusals in Article 695-24 of the Code of Criminal Procedure are where a Paris defence in a ransomware case would actually live. Surrender “peut être refusée”, the text says, in five situations: “1° Si, pour les faits faisant l’objet du mandat d’arrêt, la personne recherchée fait l’objet de poursuites devant les juridictions françaises ou si celles-ci ont décidé de ne pas engager les poursuites ou d’y mettre fin”; where a French national or resident could serve the sentence in France; “3° Si les faits pour lesquels il a été émis ont été commis, en tout ou en partie, sur le territoire français”; where France does not punish extraterritorial conduct of that kind; and where a third State already judged the same facts. Apply this to a KillSec suspect arrested in Paris. If the French prosecution has opened its own case over the French victims’ encrypted servers — and with nearly a thousand attacks worldwide, French victims are plausible — ground one lets Paris keep the suspect for the French trial. If part of the hacking was routed through infrastructure or victims located in France, ground three covers the territorial link. If the suspect is French or durably settled in France and the warrant seeks enforcement of a sentence, ground two opens the door to serving it at home. None of these grounds works automatically; each must be proven with documents, which is exactly why the Rennes litigation over Article 695-24, 2° went up to the Cour de cassation.

Age changes the procedure without blocking it. A sixteen-year-old can be surrendered under a European warrant, but French juvenile criminal law still frames everything around discernment. Article L11-1 of the Code of Juvenile Criminal Justice states: “Lorsqu’ils sont capables de discernement, les mineurs, au sens de l’article 388 du code civil, sont pénalement responsables des crimes, délits ou contraventions dont ils sont reconnus coupables.” It adds that “Les mineurs de moins de treize ans sont présumés ne pas être capables de discernement” while “Les mineurs âgés d’au moins treize ans sont présumés être capables de discernement”, defining discernment as the child having “compris et voulu son acte” and being able to grasp the meaning of the criminal proceedings. Running a ransomware affiliate panel, negotiating ransoms over encrypted messaging and laundering payments in cryptocurrency at sixteen leaves little room to rebut that presumption, yet the issuing court and the Paris court must still verify it rather than assume it. For the developer who offended as a minor and turned eighteen during the investigation, liability attaches to the age at the time of each act, so the file must date every intrusion, every exfiltration and every payment individually.

B. If Washington Asks Instead: Can France Extradite a Hacker to the United States?

When the requesting state sits outside the European Union, the European Arrest Warrant disappears and classic extradition takes over. The American presence in the KillSec operation — a United States Attorney’s Office and an FBI field office working alongside eight European states — makes this track concrete rather than theoretical. If the United States requested the extradition of a suspect arrested in Paris, the file would go before the Paris chambre de l’instruction for an opinion, and then to the French government for the final decision. Readers who want the American side of this comparison can read our analysis of the Kim Dotcom Megaupload extradition file and what France would do with a cyber-extradition of that scale, which walks through the same nationality and sentencing filters applied to a hacking-for-profit empire.

The threshold filter is nationality, and it is absolute. Article 696-4 of the Code of Criminal Procedure opens with the words “L’extradition n’est pas accordée”, followed by six bars. The first reads: “1° Lorsque la personne réclamée a la nationalité française, cette dernière étant appréciée à l’époque de l’infraction pour laquelle l’extradition est requise”. A French KillSec suspect is therefore never extradited to the United States, whatever the scale of the alleged ransoms. France would prosecute him at home instead. The rule is assessed at the time of the offence, so a naturalisation obtained after the attacks would not retroactively build a shield, and symmetrically a French national at the time of the hacks keeps the shield even if he later loses that nationality. For a sixteen-year-old operator holding French nationality, the American request stops here.

Two further bars matter for ransomware. The third provides, still under Article 696-4 of the Code of Criminal Procedure, that extradition is refused “3° Lorsque les crimes ou délits ont été commis sur le territoire de la République”. Ransomware operations are technically cross-border by design: the operator may sit in one country, the command servers in a second, the victims in a third. If any constituent act — administering the panel from a Paris apartment, exfiltrating a French hospital’s data, cashing out through a French exchange account — was committed on French soil, France claims jurisdiction and refuses to surrender. The fifth bar adds time as a weapon: refusal applies “5° Lorsque, d’après la loi de l’Etat requérant ou la loi française, la prescription de l’action s’est trouvée acquise antérieurement à la demande d’extradition, ou la prescription de la peine antérieurement à l’arrestation de la personne réclamée et d’une façon générale toutes les fois que l’action publique de l’Etat requérant est éteinte”. KillSec has operated since 2024, so limitation is not yet a live issue, but in longer-running cyber files the defence always audits both clocks, American and French, because either one running out kills the request.

Procedure then decides the suspect’s daily life while the request is pending, and a recent Cour de cassation ruling shows how fiercely that interim phase is fought. On 9 August 2023 the Criminal Chamber decided appeal no. 23-83.192 in an extradition case brought at Ukraine’s request for aggravated embezzlement and money laundering (full decision on the Cour de cassation website). The procedural snapshot is instructive: “Par arrêt du 30 mars 2023, la chambre de l’instruction a émis un avis défavorable à son extradition”, the prosecutor appealed, and the suspect then asked for his judicial supervision to be lifted, arguing that supervision “prend fin lorsque cette dernière délivre un avis défavorable et ce dès le prononcé de cet avis”. The Court’s published reasoning turns on Articles 696-11 and 696-17 and on whether the prosecutor’s appeal suspends the effects of the unfavourable opinion. Transposed to a Paris hacker wanted by Washington, the message is that an unfavourable opinion from the chambre de l’instruction is a powerful but technically fragile victory: it points toward release, yet the prosecutor’s appeal keeps the file — and potentially the supervision measures — alive until the Cour de cassation has spoken. Defence teams therefore prepare the release argument and the appeal-proof record at the same time.

Two structural differences with the warrant track deserve emphasis. First, the American request must satisfy double criminality in substance: the hacking, extortion and laundering described by the indictment must also be punishable under French law, which the next part of this article demonstrates almost always holds for ransomware, given the breadth of the French computer-crime and extortion provisions. Second, specialty and the death-penalty or inhuman-sentence guarantees that French courts routinely attach to extraditions to retentionist or harsh-sentencing jurisdictions apply here in principle, though ransomware sentences, however heavy in the United States, do not trigger the same bars as capital cases. The practical upshot for a KillSec suspect in Paris is a fork: European warrant, fast judicial track with the Article 695-24 optional refusals as the battlefield; American request, nationality and territoriality shields first, then a political decision by the government that remains reviewable by the courts.

II. From Ransom Notes to French Charges: What a KillSec File Becomes in a Paris Courtroom

Suppose France keeps the suspect — because he is French, because French courts are already prosecuting, or because part of the attacks hit French territory. The file must then be translated into French criminal charges, and the mountains of seized data must survive adversarial testing. Both exercises are demanding, and both are where French cases on computer crime are won or lost.

A. Which French Offences Does a Ransomware Business Commit, and What Prison Terms Follow?

A ransomware operation like KillSec stacks at least four layers of offending: breaking into systems, stealing and handling the data, extorting the victims, and laundering the proceeds. French law punishes each layer separately, and sentences accumulate across the stack.

The entry point is fraudulent access. Article 323-1 of the Criminal Code provides: “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.” The same article raises the penalty: “Lorsqu’il en est résulté soit la suppression ou la modification de données contenues dans le système, soit une altération du fonctionnement de ce système, la peine est de cinq ans d’emprisonnement et de 150 000 € d’amende.” KillSec’s method — exploiting poorly secured entry points, notably connected to cloud storage, in Eurojust’s description — fits the first paragraph exactly, and every deployment of an encryptor that alters the victim system’s functioning pushes the file into the aggravated second paragraph. Disrupting operations is punished on its own: Article 323-2 of the Criminal Code states that “Le fait d’entraver ou de fausser le fonctionnement d’un système de traitement automatisé de données est puni de cinq ans d’emprisonnement et de 150 000 € d’amende.” And the data theft itself is covered by Article 323-3 of the Criminal Code: “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 est puni de cinq ans d’emprisonnement et de 150 000 € d’amende.” Copying a victim’s files to the group’s own infrastructure, sending samples to prove possession, and publishing stolen data for unrestricted download when the victim refuses to pay each sit inside this single sentence several times over.

The Cour de cassation polices the boundary of these provisions strictly, which cuts both ways for the defence. On 8 June 2021 the Criminal Chamber rejected appeal no. 20-85.853 against a conviction for attacks on an automated data system: “REJET du pourvoi formé par M. [P] [B] contre l’arrêt de la cour d’appel d’Agen, chambre correctionnelle, en date du 10 septembre 2020, qui, pour atteinte à un système de traitement automatisé de données, l’a condamné à 15 000 euros d’amende et à six mois d’interdiction professionnelle” (full decision on the Cour de cassation website). The plea in that case argued that “les atteintes aux systèmes de traitement automatisé de données prévues aux articles 323-1 à 323-3 du code pénal ne sauraient être reprochées à la personne qui, bénéficiant des droits d’accès et de modification des données, procède à des suppressions de données”. The argument failed on its facts, but its structure is the template every ransomware defence in Paris will copy: contest fraudulent character access by access, log by log, credential by credential, because Articles 323-1 to 323-3 all require fraud, and authorised access used excessively is litigated differently from intrusion from the outside. Against a KillSec administrator who allegedly ran the whole panel, that line is thin; against an affiliate who merely bought access or a negotiator who only typed messages, it can be the whole case.

The ransom demand is the second layer, and French law offers two overlapping labels. Article 312-1 of the Criminal Code defines extortion: “L’extorsion est le fait d’obtenir par violence, menace de violences ou contrainte soit une signature, un engagement ou une renonciation, soit la révélation d’un secret, soit la remise de fonds, de valeurs ou d’un bien quelconque. L’extorsion est punie de sept ans d’emprisonnement et de 100 000 euros d’amende.” Threatening to publish a company’s stolen files unless it pays in cryptocurrency is the textbook illustration: obtaining funds by constraint. Article 312-10 of the Criminal Code adds blackmail: “Le chantage est le fait d’obtenir, en menaçant de révéler ou d’imputer des faits de nature à porter atteinte à l’honneur ou à la considération, soit une signature, un engagement ou une renonciation, soit la révélation d’un secret, soit la remise de fonds, de valeurs ou d’un bien quelconque.” Prosecutors routinely charge both and let the court requalify, because the threat to disclose stolen data sits at the intersection of the two definitions. The third layer is the money. Article 324-1 of the Criminal Code punishes laundering: “Le blanchiment est le fait de faciliter, par tout moyen, la justification mensongère de l’origine des biens ou des revenus de l’auteur d’un crime ou d’un délit ayant procuré à celui-ci un profit direct ou indirect”, adding that “Constitue également un blanchiment le fait d’apporter un concours à une opération de placement, de dissimulation ou de conversion du produit direct ou indirect d’un crime ou d’un délit”, with five years’ imprisonment and a 375,000 euro fine. Every hop of ransom cryptocurrency through mixers, chain-swaps and exchange accounts is a potential separate count. The fourth layer is organisation itself: Article 132-71 of the Criminal Code defines an organised gang as “tout groupement formé ou toute entente établie en vue de la préparation, caractérisée par un ou plusieurs faits matériels, d’une ou de plusieurs infractions”, and that finding aggravates most of the underlying offences while opening the door to the extended powers of organised-crime procedure. Eurojust’s description of distinct roles — “administrator, developer, negotiator and affiliate” — communicating through “encrypted messaging services” under concealing “aliases” is precisely the kind of structured, role-sharing evidence French courts use to establish the gangfinding.

Juvenile sentencing closes the picture. The sixteen-year-old operator faces the full catalogue of charges, because discernment at sixteen in a technically sophisticated, profit-driven scheme is presumed and nearly impossible to rebut, but the penalties are those of juvenile law, with educational measures taking priority and custodial sentences reduced and adapted. Article L121-4 of the Code of Juvenile Criminal Justice illustrates the scale, allowing the children’s judge to impose on a minor of at least thirteen, among other penalties, “1° De confiscation de l’objet ayant servi à commettre l’infraction”, a provision that directly threatens the seized computers, servers and crypto wallets. For the developer who offended as a minor and has since turned eighteen, each act is judged under the law applicable at the age it was committed, which is why investigators will date every intrusion individually rather than treating the two-year campaign as one block.

B. Seized Servers, 110 Terabytes and the Crypto Trail: Would the Evidence Survive in Paris?

The operation’s material harvest is spectacular: three arrests, eight house searches across Spain, Greece, the United Kingdom and Romania, five servers seized, the group’s domains taken over, and “at least 110 terabytes of stolen data” secured. Europol, for its part, supplied dedicated help to follow the cryptocurrency and analyse the digital evidence. In a French courtroom, each of these lines must pass three filters: was the seizure lawful, is the copy faithful, and does the money trail identify this defendant rather than a pseudonym?

French search-and-seizure law starts from a simple principle. Article 56 of the Code of Criminal Procedure provides: “Si la nature du crime est telle que la preuve en puisse être acquise par la saisie des papiers, documents, données informatiques ou autres objets en la possession des personnes qui paraissent avoir participé au crime ou détenir des pièces, informations ou objets relatifs aux faits incriminés, l’officier de police judiciaire se transporte sans désemparer au domicile de ces derniers pour y procéder à une perquisition dont il dresse procès-verbal.” Computer data is therefore seizable on the same footing as paper, and the same article extends searches to any premises where confiscable assets may be found. In a ransomware case the first defence audit targets the warrants and the minutes: which judge authorised each search, what time it started and ended, who was present, which devices were sealed, and whether the 110 terabytes now brandished in court are bit-for-bit images of the seized media or working copies whose handling must be documented operation by operation. A single gap in the sealing log can neutralise a whole server.

The cross-border dimension adds a second audit layer. Eurojust did not merely exchange information here: it hosted a joint investigation team between Belgium, Germany, Greece and Romania, ran the action day from its own coordination centre, and aligned five more partners — Finland, Spain, Switzerland, the United Kingdom and the United States — around shared suspect identification, infrastructure location and financial tracking. That architecture is the operation’s strength and, paradoxically, the defence’s richest terrain. Evidence gathered abroad enters the French file through mutual legal assistance and the joint-team framework, which means each transfer must carry its legal basis with it. Paris judges verify who ordered the Spanish house search signed off by Barcelona’s Investigating Court no. 20, under which regime the Swiss Attorney General’s data was transmitted, and whether the American-provided blockchain analysis was shared as intelligence or as evidence subject to adversarial debate. The distinction matters: intelligence can orient an investigation, but only evidence collected or transmitted through proper channels can ground a conviction.

Cryptocurrency tracing deserves its own paragraph because it will carry the profit motive of the whole case. Europol investigators followed the ransoms from victims’ wallets to the group’s infrastructure, and Eurojust confirms that financial trails were followed jointly by all nine authorities. In Paris, blockchain attribution is never self-proving. The prosecution must connect each wallet to a keystroke by this defendant: exchange records with identity verification, IP logs correlated with connection times, device identifiers found on seized hardware, messages in which the suspect gives deposit addresses to victims. Mixers and chain-hopping complicate the graph but do not erase it, and French courts have grown comfortable with expert testimony that walks the panel from address to address. The defence counters with the oldest argument in cybercrime — shared credentials, compromised machines, rented infrastructure — and tests it against the seized devices: if the private keys sit on the suspect’s laptop next to the administrator aliases, the argument dies; if the keys were never found, reasonable doubt breathes.

The seized data cuts both ways, and experienced Paris practitioners know the victims’ files can help the defence as much as the prosecution. Eurojust states that The authorities will now press on by studying the seized hardware and data and by following the money, which may bring to light further victims, further attacks and further participants. Every new victim identified from the 110 terabytes potentially adds counts, but every inconsistency — timestamps that contradict the alleged schedule, chat logs showing the suspect disobeying the real administrator, samples proving someone else held the data — becomes defence material through mandatory disclosure. The sheer volume guarantees years of expert work: decrypting, indexing, attributing and dating. Filing an indictment before that work is finished is the classic prosecutorial temptation, and challenging premature attribution is the classic defence response.

Conclusion

The KillSec takedown announced on 1 October 2026 is a textbook of modern cross-border prosecution: a joint investigation team, a coordination centre, nine countries, seized servers and a crypto trail. Transposed to Paris, the file would split along the lines this article has drawn. A European warrant against the teenage suspects could be executed, because sixteen is old enough for surrender and for criminal responsibility, with discernment presumed; but Paris could still refuse where French courts are already prosecuting, where the acts were committed even partly on French soil, or where the suspect would serve the sentence in France. An American extradition request would stop dead against a French national, and would otherwise be tested on territoriality, limitation and the interim battle over detention that the Cour de cassation mapped in its 9 August 2023 ruling. On the merits, French law needs no new offence to punish ransomware: fraudulent access and data extraction, extortion or blackmail over the ransom, laundering of the crypto proceeds and the organised-gang finding together cover the whole business model, with sentences that accumulate across the stack. Three reserves close the analysis, as they must. No court has established anyone’s guilt, the Eurojust investigation continues and may identify further suspects, and every contract, every log and every wallet attribution in this file remains to be debated contradictorily before a judge. That debate — over ages, over borders and over terabytes — is where cases like KillSec are truly won or lost.

📄 Official circular