In the Public Interest: The Kniaziev Case as an Indicator of a Systemic Problem with Plea Agreements

In May 2026, former President of the Supreme Court (SC) Vsevolod Kniaziev admitted his guilt in accepting a bribe of almost three million US dollars three years earlier. He entered into a plea agreement with a prosecutor of the Specialised Anti-Corruption Prosecutor’s Office (SAPO) and provided purportedly incriminating testimony concerning other Supreme Court judges, which is expected to form the basis for their conviction for corruption offences. The High Anti-Corruption Court (HACC) sentenced Kniaziev to five years’ imprisonment, barred him from holding positions in judicial bodies for a further three years after serving the principal sentence and confiscated his property. In addition, the convicted person donated one million US dollars to support the Armed Forces of Ukraine through a non-governmental charitable foundation.

But was the agreement concluded in the public interest and does such an outcome amount to ‘buying one’s way out of justice’? How can someone donate an undeclared one million US dollars without raising questions from law enforcement authorities? In this article, we examine the systemic problems in the practical application of plea agreements in criminal proceedings that the Kniaziev case has brought to light.

Vsevolod Kniaziev joined the Supreme Court from a first-instance court through the 2016–2017 competition. At the time of the competition, there were no negative opinions concerning him from civil society, and at the interview he presented himself as a young and promising judge. This subsequently helped him first become Secretary of the Grand Chamber of the SC and later, in contrast to candidates from the ‘old system’, become President of the Supreme Court at the age of 42 with the support of a majority of SC judges. A position that a judge with less than ten years of experience could normally only dream of.

In public, Kniaziev spoke about the need to combat corruption in the judiciary and, at international meetings, expressed zero tolerance for ‘telephone justice’, cronyism and corruption. As later became known, at the same time the President of the SC was building his own ‘vertical of power’ and, according to the authors of this article, meeting informally with the Head of the Office of the President of Ukraine Andriy Yermak. While in spring 2022 loyalty to the Presidential Office affected only the transfer of some judges from the regions to Kyiv courts (for example, Judge Svitlana Hrechana), a year later the same dynamic led Kniaziev to cross all boundaries. He then decided to accept the largest bribe in the history of the judiciary for a decision of the Grand Chamber of the Supreme Court that was lawful on its merits.

On the evening of 15 May 2023, the media erupted with reports that the President of the SC had been exposed while taking a USD 2.7 million bribe. It was reported that the National Anti-Corruption Bureau of Ukraine (NABU) was conducting searches at the Supreme Court itself, at the premises of all judges of the Grand Chamber, at the premises of some SC staff members and directly at Kniaziev’s home.

At the time he was served with a notice of suspicion, Kniaziev was the biggest ‘fish’ in the justice system ever caught by the NABU and SAPO. At a press conference concerning his detention, the heads of the anti-corruption bodies also stated that corruption in the SC was extensive. According to them, other judges of the Grand Chamber were allegedly involved, as the ‘Kniaziev tapes’ recorded him instructing that envelopes be prepared for judges of the Grand Chamber of the SC. At the time, it seemed that this concerned virtually everyone in the Grand Chamber (21 judges), and the public therefore expected further notices of suspicion and convictions. This did not happen.

The following day, the Plenum of the SC removed Vsevolod Kniaziev as President, the High Council of Justice (HCJ) gave its consent to his detention in custody and Kniaziev himself spent 8.5 months in a pre-trial detention centre, until his ‘fairy-tale friends’ posted more than UAH 18 million in bail for him.

The investigation lasted slightly less than ten months, which is entirely normal for a case of this category. One month after Kniaziev was released from the pre-trial detention centre, the indictment against him was submitted to the High Anti-Corruption Court. The HACC then heard the case almost every week until the plea agreement was concluded in June 2026. Shortly before that, in May 2026, four former judges of the Grand Chamber of the SC were served with notices of suspicion on the basis of Kniaziev’s testimony. The investigation concerning them is ongoing.

Three of them challenged the HACC decision approving Kniaziev’s agreement with the prosecutor before the Appeal Chamber of the HACC, but the Chamber refused to consider their appeals. The Criminal Cassation Court of the Supreme Court alone can now return the Kniaziev case to the Appeal Chamber of the HACC if it finds grounds to do so — provided, of course, that the decision of the Appeal Chamber of the HACC is appealed to the Criminal Cassation Court of the Supreme Court.

Under the HACC judgment, Kniaziev was found guilty of the criminal offence under Part 4 Article 368 of the CC of Ukraine for accepting an unlawful benefit in exchange for a court decision. Under the terms of the agreement, his sentence is more lenient than under the general rule (up to 12 years’ imprisonment), because Kniaziev purportedly provided testimony that is expected to lead to the conviction of other judges for corruption offences.

The sentence is five years’ imprisonment, of which Kniaziev will in practice serve around four years, taking into account the time he has already spent in the pre-trial detention centre. He is also barred for three years from holding positions in courts and law enforcement bodies (an additional penalty, the term of which will begin after he has served the principal sentence). The court also confiscated Kniaziev’s property: an apartment, a house and funds amounting to UAH 32,000, USD 201,000 and EUR 1,700. The funds seized during the search were subject to special confiscation because they constituted the unlawful benefit, that is, the funds used to pay the bribe.

Charitable giving is an additional condition of Kniaziev’s agreement. He transferred USD 1,104,600 to the Come Back Alive Foundation. These funds had been seized during searches of Kniaziev and held as material evidence in the case.

A plea agreement is an alternative means of resolving a criminal-law conflict and a mechanism for accelerating criminal proceedings. It helps optimise procedural time, simplify the consideration of cases, reduce the workload on pre-trial investigation bodies and the court and lower procedural costs. The resources freed up can then be used to uncover other crimes.

The logic is straightforward: a prosecutor may enter into an agreement with a suspect or accused person at any stage of criminal proceedings. The prosecution saves resources, while the defence receives a predictable, agreed sentence — in exceptional cases, even below the minimum threshold provided for by the sanction of the relevant article.

The HACC has actively used plea agreements since it began operating: over six years, half of its judgments have been based on them. The trend became particularly striking after November 2024, when Law No. 4033-IX entered into force. Before then, the share of plea agreements at the HACC ranged between 25% and 40%. Since then, two thirds of convictions have been based on agreements. ALI experts have written about this instrument in greater detail before.

This does indeed save resources in investigations where suspects do not admit guilt, convincing evidence sufficient for conviction is difficult to obtain and the 10- to 15-year limitation periods for criminal liability are, in some cases, approaching expiry.

The key condition for entering into a plea agreement is the existence of a public interest in doing so. In other words, the prosecutor and the judge must be satisfied that society has an interest in expediting proceedings and uncovering a greater number of criminal offences. Entering into an agreement is a right of the suspect or accused person: the prosecutor is not obliged to conclude it and the judge is not obliged to approve it if there is no public interest. It is therefore not a universal instrument for every situation and cannot replace effective work by the prosecution. When the practice becomes ubiquitous, the value of the public-interest rationale is lost and the very logic of plea agreements is undermined.

At the same time, when assessing whether a public interest exists, the HACC relies only on the materials provided to it by the parties to the criminal proceedings in connection with the agreement. The court’s discretion is therefore extremely constrained by the information provided, and it is not surprising that judges accept these arguments. In addition, the procedural consequence of refusing to approve an agreement is unclear: the trial then proceeds under the ordinary procedure. This is extremely difficult where the person has admitted guilt, disclosed evidence and provided information concerning other persons, among other things.

So how does the HACC substantiate the existence of a public interest specifically in the Kniaziev case?

The HACC states that ‘a real sentence is the important component of this agreement that underscores the public interest, the inevitability of punishment and the restoration of trust in justice’. Most plea agreements involve a suspended prison sentence and therefore often look, in the eyes of the public, like ‘buying one’s way out of justice’. It is worth noting that lawyer Oleh Horetskyi, who described himself as ‘Kniaziev’s back office’, received precisely such an agreement from SAPO and the HACC (a five-year suspended sentence and a UAH 21 million donation).

The judges also considered that concluding the agreement with the former judge substantially saved prosecution resources that would otherwise have been spent on further examination of evidence. In this way, Kniaziev helped free up resources for investigating other corruption offences. The HACC states that his admission of guilt resulted in ‘a reduction in the workload of the prosecution service, the Bar and the courts, which gain more time to uncover other crimes and consider them in court…’.

At the same time, SAPO entered into the agreement with Kniaziev only at the trial stage. By then, around fifty court hearings had already taken place and considerable resources had been spent on gathering and examining evidence. One might expect that two years of trial proceedings would indicate a convincing prosecution case ready to be taken through to a conviction rather than abandoned in favour of an agreement. It therefore remains unclear why the agreement was not concluded at the early stages of the pre-trial investigation, as is usually the case. Here, the volume of resources saved is substantially smaller than in other proceedings.

Another component of the public interest is exposing other persons. This part of the judgment is not publicly available, as it is not subject to disclosure by law. Most likely, in 2026 Kniaziev was used as a ‘torpedo’ — to communicate with SC judges about the circumstances of the case. Recordings of these conversations may be used to prove their guilt. On the basis of this testimony, the NABU subsequently served notices of suspicion on other persons involved in the case. According to the investigation, Supreme Court judges Iryna Hryhorieva, Zhanna Yelenina and Ihor Zhelieznyi, as well as retired SC judge Oleksandr Prokopenko, received an unlawful benefit for delivering a decision in Zhevaho’s interests. There is currently little official information on the progress of these criminal proceedings. The question naturally remains: what could the judges have said in a conversation with Kniaziev three years later, while his case was already being heard in court? It is therefore impossible to assess how significant the information provided by Kniaziev will be for the subsequent evidentiary process leading to convictions.

SAPO has publicly stated that ‘thanks to the agreement, the State obtained a substantial economic benefit: the country’s budget received USD 2,554,300. That is more than UAH 113 million! Today, as the country fights for its survival, these funds are converted into real weapons, drones or financing for social payments’. On the one hand, this is a strong message, as the funds had remained for years in frozen accounts and in storage as material evidence. They have now been channelled into the economy and are being spent directly on national defence.

On the other hand, the charitable contribution to the Armed Forces — more than USD 1.1 million — raises a legitimate question: where did this money come from? It was not declared in Kniaziev’s asset declaration. Taking joint ownership into account, he could legally have transferred an amount ten times smaller. Yet under the agreement these funds have already been confiscated in favour of the State. The origin of the USD 1.1 million should therefore have attracted the attention of both the financial monitoring authorities when the funds were transferred to the charitable foundation’s account and the anti-corruption bodies.

Against the background of this judgment, the asset declaration issue may of course appear minor, but the absence of communication on this point looks like the ‘legalisation’ of proceeds of crime.

Personal funds are also transferred in support of the Armed Forces under other plea agreements, but what is their origin? Does this agreement not call into question the origin of the funds transferred by most persons convicted on the basis of plea agreements?

Support for the Armed Forces, including through charitable foundations, is nevertheless a private matter for each person. It is not a mandatory condition of a plea agreement but rather a circumstance that reflects positively on the person. At the same time, the CPC of Ukraine does not require optional elements of an agreement to be stated in the court judgment, so the donation to the Armed Forces should not have appeared there at all.

The idea of plea agreements as a way to maximise the punishment of corrupt offenders — tackling not an isolated phenomenon but the system — is sound.

The Kniaziev case may appear typical of this practice, especially as regards the ‘donation to the Armed Forces’. At the same time, it is not. The HACC very rarely imposes a sentence of actual imprisonment under a plea agreement. This is related both to the high-profile nature of the case and to the fact that the proceedings were already moving towards a judgment, having been before the court for more than two years.

At the same time, the case exposed systemic problems in the practice of concluding plea agreements. However much Ukrainian society may trust the anti-corruption bodies, there are widespread, well-founded doubts about the policy of routinely concluding plea agreements.

The criteria used by SAPO to conclude agreements. They are currently non-transparent. In the public communications of SAPO and the HACC, no theoretical rationale is offered for plea agreements. The rationale concerns either a person’s rational choice or the maximisation of criminal prosecution by exposing a greater number of corrupt offenders. This needs to be systematised and developed into a separate publicly articulated SAPO policy.

The concept of ‘public interest’. It remains evaluative and ambiguous. It can be elaborated both through HACC practice and through the Supreme Court’s quasi-precedential positions (legal positions). In addition, the legislator could define the public-interest criterion more clearly in the CPC of Ukraine.

Uncovering other criminal offences and persons involved must yield real results, meaning that it must lead to convictions rather than merely create a prospect of them. It is understandable that this sensitive information cannot be disclosed immediately in the judgment, since it will subsequently be used to prove other criminal offences. But this is precisely what most undermines trust in plea agreements: the public does not see tangible results. How many people have been convicted on the basis of incriminating testimony, who were they, how and when were they convicted? Such information can be provided later and in aggregated form.

More radical steps would be to require SAPO and the HACC to publish the full text of a judgment of conviction based on a plea agreement once the incriminating information has been used in court. Alternatively, publication could become mandatory three years after the agreement is concluded, so that the use of the incriminating information can be monitored.

A donation to the Armed Forces of Ukraine cannot form part of a plea agreement. Support for the Armed Forces, including through charitable foundations, is a private matter for each person. Charitable giving should under no circumstances be mentioned in a court judgment, so as not to create a sense of impunity — as if the convicted person had ‘bought their way out of punishment’.

A substantive analysis of plea agreements should form part of the audit of the Specialised Anti-Corruption Prosecutor’s Office. Pursuant to Part 10 Article 8-1 of the Law of Ukraine ‘On the Public Prosecutor’s Office’, SAPO undergoes an audit of its activities every two years. In practice, no such audit has yet taken place, although this provision was introduced in 2023. The audit is conducted by the Commission for External Independent Assessment, which includes international experts. The criteria mentioned above, their practical implementation at the level of criminal prosecution policy and the prospects of convicting persons exposed as a result of plea agreements — all of this should be examined through periodic audits as an instrument of external oversight of SAPO’s activities. SAPO itself also has an interest in the results of regular audits: they provide a credible response to critics or create space for development and for addressing mistakes.

If SAPO strengthens trust in plea agreements, this instrument can be preserved and used to reduce corruption in the country. If it continues its current practice, trust in the NABU, SAPO and the HACC will decline, and under slogans such as ‘they let corrupt offenders buy their way out’, MPs, supported by an outraged public, may restrict these powers.

At present, everything rests on public trust in SAPO and the HACC. In other words, on the assumption that prosecutors act in good faith in the interests of justice and enter into an agreement, or initiate one themselves, only where society genuinely benefits. Accordingly, they have to choose between the risk of ‘losing the case’ and investigating it for so long that the same resources would be better directed to other cases. After all, fair justice is utilitarian: it lies not in imposing a harsh punishment on one person at any cost, but in ensuring the inevitability of punishment for everyone — perhaps a more lenient punishment, but punishment nonetheless. This trust will, however, erode if agreements remain non-transparent and closed, sentences are mostly suspended and the criteria for concluding agreements remain unclear.