On 26 August 2026, Reuters reported that the British-based chief executive of a cryptocurrency company who is wanted in the United States on charges of wire fraud and market manipulation faces being sent to the United States to stand trial after losing his fight against extradition the previous week. The man is Manpreet Kohli, a 45-year-old Indian national, named among the defendants in the Justice Department operation against widespread cryptocurrency fraud and described as the leader of Saitama, a company that promoted a token on the ethereum blockchain which at one point had a market value of 7.5 billion dollars. American prosecutors allege that Kohli and his co-conspirators claimed to be holding and purchasing Saitama tokens while privately selling them for millions of dollars in profit, and that Kohli himself made around 20 million dollars. He was arrested in London in 2024 after United States prosecutors accused him and more than a dozen others of widespread fraud, following an investigation in which the FBI for the first time set up a digital token to ferret out crime. On 19 August 2026, Judge Samuel Goozee rejected his challenge, finding that American custody could cope with his mental health, and sent the case to British ministers for the final decision. Kohli remains at liberty on 200,000 pounds bail and can still appeal. The question for an international reader with ties to France is direct: if the same man had been stopped at Roissy or in the streets of Paris, would France have handed him to America, and how would a Paris lawyer have fought the file? This analysis answers with the exact French procedure, the governing texts, and four recent rulings of the Cour de cassation.
I. Would France surrender a crypto defendant to the United States?
A. Wire fraud in America, fraud in France: can a Saitama-type file pass the double-criminality test?
The first question a Paris extradition lawyer asks when an American request lands on the desk is whether the facts described by the requesting state are also punishable under French criminal law. French extradition outside the European arrest warrant system is governed by a chapter of the Code of Criminal Procedure which applies whenever an international convention does not provide otherwise, since article 696 of the Code of Criminal Procedure provides that “En l’absence de convention internationale en stipulant autrement, les conditions, la procédure et les effets de l’extradition sont déterminés par les dispositions du présent chapitre.” Within that chapter, article 696-3 of the Code of Criminal Procedure sets the double-criminality threshold in plain terms: “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.” A fraud file that exists only under American law cannot cross the Atlantic in the other direction.
Applied to the Kohli file, that test would be met without serious difficulty. The American charges are wire fraud and market manipulation built on a classic allegation: public claims of holding and buying a token while secretly selling it for about 20 million dollars of personal profit, inside a scheme whose token once capitalised 7.5 billion dollars. Under French law, conduct of that shape falls within fraud and related financial offences punishable as correctional or criminal penalties, so the double-criminality condition of article 696-3 would be satisfied on the papers. The defence would therefore gain nothing by arguing that wire fraud is an American doctrinal invention with no French equivalent. French courts do not demand that the foreign qualification have a twin with the same name in the French codes. They check that the underlying acts, as described, amount to an offence under French law.
That method is exactly what the Cour de cassation restated in a European arrest warrant case that Paris extradition lawyers cite in every double-criminality discussion. In its judgment of 11 August 2021, No. 21-84.361, published at Cass. crim., 11 August 2021, No. 21-84.361, the criminal chamber held that “les juges doivent seulement s’assurer que les faits à l’origine du mandat d’arrêt constituent une infraction au regard de la loi pénale française.” The judges need only verify that the acts behind the warrant are an offence in the eyes of French criminal law. The same reasoning governs classical extradition through article 696-3: the chamber of the investigation division maps facts to French offences; it does not translate American statutes word for word. A pump-and-dump presentation of a token, coordinated selling behind investors’ backs, and millions in concealed profit would map onto French fraud without strain, and the FBI’s method, setting up a digital token for the first time to detect the crime, however striking for the press, changes nothing in that legal mapping. Investigative originality does not create or destroy double criminality.
The most useful precedent for a Kohli-type file is only a few months old and comes from a French court dealing with an American request. On 14 April 2026, the Cour de cassation rejected a challenge against a favourable opinion given by the investigation chamber of the Aix-en-Provence Court of Appeal to a United States extradition request, in Cass. crim., 14 April 2026, No. 25-86.772. The request had been issued “en exécution de mandats d’arrêt délivrés pour l’exercice de poursuites des chefs d’association de malfaiteurs en vue de commettre une escroquerie en valeurs mobilières et fraude électronique, escroquerie en valeurs immobilières et fraude électronique, aux Etats-Unis d’Amérique, courant janvier 2009 jusqu’à septembre 2014”, and the formal extradition request had been notified on 30 April to a man who, the Court records, did not consent: “Une demande d’extradition a été notifiée le 30 avril suivant à M. [U], qui n’a pas consenti à son extradition.” Securities fraud, wire fraud, criminal association aimed at fraud: the Aix-en-Provence chamber gave a favourable opinion, and the Cour de cassation rejected the appeal. For a crypto defendant, the message is blunt. An American financial-fraud file of the Saitama type, wire fraud plus market manipulation over several years, is precisely the kind of request French investigation chambers already approve when the papers are in order. The fight must therefore move to the other conditions: who the person is, what penalty threatens them, and whether the procedure that awaits them meets fundamental guarantees.
B. French nationality, death penalty and unfair trial: the three red lines that can stop a surrender to America?
Once double criminality is established, French law draws a series of absolute prohibitions that no American request can cross. They are listed in article 696-4 of the Code of Criminal Procedure, which opens with the words “L’extradition n’est pas accordée : 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”. France does not extradite its own nationals to non-European Union states. Kohli is an Indian national, so this shield would not have helped him in Paris, but for the many dual nationals and French residents who read extradition news with personal anxiety, it is the first line of the analysis: a French passport at the time of the alleged acts closes the American file at the threshold, and the requesting state is then invited to transmit its evidence so that France can prosecute at home.
The same article continues with two prohibitions that matter in every American file. Extradition is refused, in the words of article 696-4 of the Code of Criminal Procedure, “6° Lorsque le fait à raison duquel l’extradition a été demandée est puni par la législation de l’Etat requérant d’une peine ou d’une mesure de sûreté contraire à l’ordre public français”, and “7° Lorsque la personne réclamée serait jugée dans l’Etat requérant par un tribunal n’assurant pas les garanties fondamentales de procédure et de protection des droits de la défense”. Wire fraud and market manipulation do not carry the death penalty, so the sixth ground would not decide a Saitama-type case. But the architecture matters because American files sometimes aggregate charges, and the defence in Paris always audits the full penalty exposure, including any count that could theoretically open a capital or irreducible punishment, as well as the reality of the trial guarantees: access to an independent judge, rights of the defence, and detention conditions that comply with Article 3 of the European Convention on Human Rights.
On the death-penalty point, the Cour de cassation has set a demanding standard that Paris lawyers invoke whenever an assurance from the requesting state looks vague. In Cass. crim., 30 January 2024, No. 23-83.549, the criminal chamber recalled that “Le premier de ces textes prohibe l’extradition vers un autre État d’une personne dont il y a des motifs sérieux et avérés de croire qu’elle serait exposée dans le pays de destination à un risque réel d’être soumise à la peine de mort.” It then quashed a favourable opinion because the investigation chamber had contented itself with general considerations about a moratorium instead of a clear commitment, holding that “lorsque la peine de mort est encourue à raison des faits pour lesquels l’extradition est demandée, il appartient à la chambre de l’instruction de s’assurer que l’Etat requérant prend l’engagement dénué de toute ambiguïté que cette peine ne sera pas appliquée à la personne réclamée.” An undertaking with no ambiguity: that is the yardstick. In a crypto case with no capital count, this line of argument stays in reserve, yet it shapes the whole defence method, because the same insistence on concrete, verifiable guarantees irrigates the Article 3 debate on prison conditions and health risks.
That health-risk debate is precisely where Kohli fought in London and lost. He argued that the United States would lack adequate measures to stop him taking his own life in custody. Judge Goozee answered on 19 August 2026 that arrangements during transit and within the American prison system could manage Kohli’s mental health and keep the suicide risk at an acceptable level. In Paris, the same argument would have been channelled through Article 3 and the seventh ground of article 696-4, with the investigation chamber examining American detention assurances, medical-care arrangements, and the personal medical file, rather than through the British forum-bar and human-rights-bar framework. The outcome cannot be predicted case by case, but the method is settled: general American assurances about the prison system do not suffice by themselves; the chamber must verify that the guarantees are precise, effective, and adapted to the person before it, as the 30 January 2024 ruling requires for the death penalty and as investigation chambers apply by extension to inhuman-treatment risks. A defendant with a documented psychiatric history therefore has a genuine, if difficult, path in Paris: expert reports, a detailed American detention plan, and a reasoned demonstration that the file’s general promises do not cover their specific vulnerability.
A final red line deserves mention because American crypto files often involve defendants already convicted or pursued elsewhere. French law refuses surrender where the acts were committed on French territory, where they have already been finally tried in France, or where limitation has extinguished the prosecution, and it refuses surrender for political or military offences. None of these fits a Saitama-type pump-and-dump, but a Paris lawyer always checks them, because crypto schemes are typically multi-jurisdictional: servers in one country, promoters in another, victims everywhere. If part of the alleged manipulation was directed from French soil, the territoriality objection and the option of a French prosecution become live arguments, and the American request must then survive a much harder examination.
II. Arrested in Paris on an American request: the procedure and the defences that actually work?
A. Diplomatic request, detention under extradition warrant and reasoned opinion: the Paris hearing step by step
In France, an American extradition begins with paper, not with handcuffs at the airport. Article 696-8 of the Code of Criminal Procedure provides that “toute demande d’extradition est adressée au gouvernement français par voie diplomatique et accompagnée soit d’un jugement ou d’un arrêt de condamnation, même par défaut, soit d’un acte de procédure pénale décrétant formellement ou opérant de plein droit le renvoi de la personne poursuivie devant la juridiction répressive, soit d’un mandat d’arrêt ou de tout autre acte ayant la même force et décerné par l’autorité judiciaire”. The request must identify precisely the acts and their date. For a crypto file, that means the American arrest warrants, the indictment describing the token, the alleged false statements, the concealed sales and the 20-million-dollar profit, with dates and places. A vague narrative about a 7.5-billion-dollar token and market excitement would not suffice; the chamber wants acts, dates, and the American judicial act that embodies them. Where the person is wanted urgently, a provisional arrest can precede the formal request, but the requesting state must then produce its papers within the applicable time limit or the detainee must be released, which is why Paris lawyers count days from the first hour.
Once the person is in custody, the general prosecutor presents them to the first president of the court of appeal, and article 696-11 of the Code of Criminal Procedure states that “Le premier président de la cour d’appel ou le magistrat du siège désigné par lui ordonne l’incarcération et le placement sous écrou extraditionnel de la personne réclamée à la maison d’arrêt du siège de la cour d’appel.” Detention under an extradition warrant is the default, at the prison of the seat of the court of appeal, in Paris the Santé or Fresnes circuit depending on allocation. Release under judicial supervision remains possible where representation at every stage of the proceedings is sufficiently guaranteed, and in a crypto case with a settled Paris residence, a family, professional roots and a surrendered passport, that argument is worth making from day one, even though investigation chambers grant it sparingly in American files where the incentive to flee is deemed high and the alleged amounts reach tens of millions.
The decisive hearing takes place before the investigation chamber of the Paris Court of Appeal. Where the requested person has told the general prosecutor that they do not consent to surrender, the chamber is seized without delay and the person appears within ten working days. Then article 696-15 of the Code of Criminal Procedure governs the outcome: “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.” A reasoned opinion, not a judgment of guilt: the chamber verifies each legal condition, refuses where a condition fails or where there is an obvious error, and delivers its opinion, absent further inquiries, within one month of the appearance. Refusing consent is therefore a strategic fork, not a mere protest. The recent case law shows defendants systematically taking it: in the Algerian file that reached the Cour de cassation on 27 May 2026, Cass. crim., 27 May 2026, No. 25-84.904, the Court notes as a matter of course that “M. [D] n’a pas consenti à sa remise aux autorités requérantes”, before examining an opinion that was partly favourable in a case involving a life sentence imposed in absentia for organised drug trafficking and money laundering. Consent would have meant a simplified, rapid surrender; non-consent opened the full adversarial examination, three investigation-chamber rulings in that file, and then review by the Cour de cassation.
The 14 April 2026 American-file ruling illustrates how seriously that adversarial examination is conducted and how the Cour de cassation polices it. In No. 25-86.772, the defence attacked the composition of the chamber and the sequencing of its rulings, arguing that the second panel had merely referred back to the first panel’s findings on limitation and fair-trial guarantees instead of verifying every condition itself. The Cour de cassation answered that each ground, even one previously addressed before further inquiries were ordered in a different composition, must be argued before the chamber and decided by it in an identical composition, and it confirmed that on the record before it the chamber had breached none of the cited texts. The lesson for a crypto defendant is twofold. First, the investigation chamber can order additional inquiries, seek American assurances, and split its examination across several rulings, as it did in both the American and Algerian files. Second, every ground must be pleaded in full before the deciding panel: limitation, double criminality, political character, penalty exposure, Article 3 risks, fair-trial guarantees. A ground raised only by reference to an earlier ruling is a ground half-defended, and extradition litigation punishes half-defended grounds.
Concretely, a Saitama-type file in Paris would be fought on four fronts at that hearing. Double criminality, mapped act by act onto French offences rather than contested as a matter of labels. Limitation, computed under both American and French law, because time extinguishes the right to surrender. The penalty and detention package, tested against public policy and Article 3 with documentary evidence, not rhetoric. And the regularity of the American papers themselves: precise acts, precise dates, a judicial act of the required force. The chamber’s opinion is unfavourable wherever a legal condition fails or an obvious error appears, and an unfavourable opinion in principle ends the French phase. A favourable opinion, as in the Aix-en-Provence American file and partly in the Algerian file, sends the case to the government for the decree stage, and to the Cour de cassation for review.
B. Release, cassation and the decree: the last remedies before the plane?
Detention under an extradition warrant is not a tunnel with no exits. Article 696-19 of the Code of Criminal Procedure provides 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”. At any time: the adverb matters. The defence can file for release on the first day and again whenever the file evolves, for example when the American papers arrive late, when an expert report documents a health condition incompatible with continued detention, or when the proceedings stretch beyond what the case requires. Kohli’s London position, at liberty on 200,000 pounds bail pending the ministerial decision and a possible appeal, has its Paris equivalent in this permanent release track, combined with the possibility of supervision measures where the first president considers that appearance at every procedural act is sufficiently guaranteed. In practice, American files make judges cautious, because the alleged profits, 20 million dollars in the Saitama narrative, suggest means and mobility. The winning release files are therefore concrete: a Paris home, a family, employment, a surrendered passport, a medical monitoring plan, and, where relevant, an offer of electronic supervision. Assertions do not release detainees; anchors do.
A favourable opinion is then open to review before the Cour de cassation, and the three recent rulings used in this analysis prove that this review is real and regularly sought: a pourvoi against the Aix-en-Provence favourable opinion in the American securities-fraud file rejected on 14 April 2026, a pourvoi against a partly favourable opinion in the Algerian file rejected on 27 May 2026, and a cassation granted on 30 January 2024 where the chamber had relied on vague assurances about the death penalty. The Cour de cassation does not re-try the facts or re-weigh the American evidence. It verifies that the investigation chamber answered every ground, gave reasons capable of supporting its opinion, and applied the correct legal standards to limitation, double criminality, public policy, and fundamental rights. The 30 January 2024 cassation is the proof that this control bites: insufficient reasoning on the reliability of foreign assurances means annulment. For a crypto defendant, the pourvoi is therefore not a formality. It is the moment when a meticulously built record, expert reports, American prison documentation, precise penalty computations, pays off, because the criminal chamber can only sanction errors that the defence has properly placed in the file below.
If the opinion is favourable and survives cassation review, the file passes to the government, since article 696-2 of the Code of Criminal Procedure provides that “Le gouvernement français peut remettre, sur leur demande, aux gouvernements étrangers, toute personne n’ayant pas la nationalité française qui, étant l’objet d’une poursuite intentée au nom de l’Etat requérant ou d’une condamnation prononcée par ses tribunaux, est trouvée sur le territoire de la République.” The surrender is ordered by an extradition decree signed by the Prime Minister on the report of the Minister of Justice. That decree is itself an administrative act open to challenge before the administrative courts, including urgent interim proceedings to suspend its execution, and, in parallel, the defence may seek interim measures from the European Court of Human Rights where surrender would expose the person to a real risk of treatment contrary to Article 3 or to a flagrant denial of justice. These final remedies run on very short clocks measured in days, sometimes hours, once the decree is notified. They must be prepared while the cassation is still pending, not improvised after it.
One clarification matters for readers fascinated by the FBI’s honeypot token. The Paris investigation chamber does not put the American investigation on trial. Whether federal agents were entitled to create a digital token to attract fraudsters, whether that amounts to entrapment under American law, and whether the resulting evidence is admissible are questions for the American trial court, not for the French chamber, whose office is limited to the legal conditions of surrender. A Paris lawyer therefore parks the honeypot debate for the American defence team and concentrates the French hearing on what Paris decides: punishability on both sides, limitation, nationality, penalties, trial guarantees, and detention risks. That division of labour is also the reason international crypto defendants need two teams from the start, one in the requesting state and one in the requested state, coordinated around a single timeline, because the French clock keeps running while the American strategy is built.
Conclusion
The Kohli case, as reported on 26 August 2026, is a textbook cross-border fraud file: a 7.5-billion-dollar token, 20 million dollars of alleged personal profit, an FBI operation that created its own bait, an arrest in London in 2024, and an extradition challenge fought and lost on 19 August 2026 over suicide-risk guarantees, with ministers now holding the pen and an appeal still available. Transposed to Paris, the same file would almost certainly clear the double-criminality threshold, as the 14 April 2026 American-file precedent shows, and would then be fought where French extradition law is most demanding: precise American papers, limitation, the penalty package, trial guarantees, and the concrete medical and detention evidence behind every Article 3 claim. The 30 January 2024 cassation stands as a warning to requesting states and an instrument for the defence: vague assurances do not survive review. Anyone reading this with an American file on their horizon, a summons, an arrest, or simply the discovery that a foreign warrant may exist, should act before the handcuffs close. Identify the requesting state and the exact charges, preserve passports and residence proofs, assemble the medical file, and instruct counsel in both countries while release, non-consent strategy, and interim remedies are still fully available. Extradition rewards the early and punishes the late.