MEDIA AND COMMUNICATIONS LIST
Royal Courts of Justice
Strand, London, WC2A 2LL
Before :
MR JUSTICE CHAMBERLAIN
Between :
MINISTRY OF DEFENCE | Claimant |
- and - | |
(1) GLOBAL MEDIA AND ENTERTAINMENT LIMITED (2) INDEPENDENT DIGITAL NEWS AND MEDIA LIMITED (3) TIMES MEDIA LIMITED (4) ASSOCIATED NEWSPAPERS LIMITED (5) TELEGRAPH MEDIA GROUP HOLDINGS (6) THE FINANCIAL TIMES LIMITED (7) THE PRESS ASSOCIATION LIMITED COMPANY (8) BARINGS LAW | Defendants Defendant for the purposes of costs only |
Cathryn McGahey KC, John Bethell and Anisa Kassamali (instructed bythe Government Legal Department) for the Ministry of Defence
Jude Bunting KC (instructed by Pia Sarma, Editorial Legal Director, Times Media Limited) for the First to Sixth Defendants
Jess Glass (Journalist) and Louise Irwin (Group General Counsel) on behalf of the Seventh Defendant
Tom Forster KC and Paul Mertens (instructed by the Special Advocates’ Support Office) as Special Advocates
Hearing dates: 1 September, 13 October, 1 November, 18 December 2023, 1-2 February, 27 March, 30 April,1 May, 11 and 15 November 2024 and 20-21 February, 9, 19 and 23 May, 26 June and 1, 4, 7, 8, 11 and 14 July 2025
APPROVED JUDGMENT
Mr Justice Chamberlain:
Introduction
This, my fourth judgment in this case, explains why a super-injunction contra mundum (against the world), granted on the application of the Ministry of Defence (“MOD”) on 1 September 2023, is now being discharged nearly two years later. My first three judgments, which were handed down in private, are also published today.
The super-injunction prevented disclosure of:
the fact of the release by someone working for the UK Government of a dataset containing personal information and contact details of persons who applied for relocation to the UK from Afghanistan following the Taliban coup in 2021; and
the existence of the injunction itself.
It was granted and continued because of the risk that, if the Taliban learned about the existence of the dataset, it is likely that they would be able to acquire it and would use it to identify those who had applied for relocation and target them for extra-judicial killing or severe physical ill-treatment.
The super-injunction is now being discharged following a review commissioned by the MOD, which concludes that the Taliban likely already possess the key information in the dataset, that it is unlikely that individuals would be targeted simply because of their work for the UK or allied governments or for the former Government of Afghanistan and that the acquisition of the dataset is accordingly unlikely substantially to raise the risk faced by the individuals whose data it includes.
Background
The dataset at issue in these proceedings contains details of those who had applied under the Afghan Relocations and Assistance Policy (“ARAP”) and the ex gratia predecessor of the Afghan Citizens’ Resettlement Scheme (“ACRS”) for relocation to the UK on the ground that they were at risk from the Taliban as a result of their work with or for the UK prior to the Taliban coup in 2021. It was released in error in early 2022. In August 2023, it became known to the MOD that part of the dataset had been published on a Facebook page. Shortly after this, the MOD applied for an injunction.
Although the MOD did not originally apply for one, the judge to whom the application was first made, Robin Knowles J, decided on 1 September 2023 to grant a super-injunction—that is to say an injunction which prohibited disclosure of the existence of the order, as well as the underlying information. Robin Knowles J’s reasons for granting that form of relief were set out in a written ruling handed down in private on 2 September 2023. His ruling is also being published today.
Neither the hearing before Robin Knowles J, nor any of the subsequent hearings, has taken place in public. At some of the hearings, representatives of the media who have been notified of the injunction have been present. The number of these representatives has gradually increased. The media organisations now involved are the entities which own The News Agents (a podcast), The Independent, The Times, The Daily Mail, The Daily Telegraph, The Financial Times and the Press Association.
There have also been CLOSED hearings from which media representatives were excluded, initially under the inherent jurisdiction (before the media organisations became parties) and then under the Justice and Security Act 2013. The interests of the media defendants have been represented at these hearings by Tom Forster KC and Paul Mertens, initially as Advocates to the Court and then as Special Advocates.
The super-injunction was continued by me after hearings on 13 October 2023, 31 October-1 November 2023, 1 and 18 December 2023, 22 January 2024 and 1-2 February 2024. I gave judgments explaining my reasons on 23 November 2023 (“Judgment no. 1” [2023] EWHC 2999 (KB)) and 15 February 2024 (“Judgment no. 2” [2024] EWHC 312 (KB)).
In my Judgment no. 1, at [35]-[39], I noted that the grant of a super-injunction in the circumstances of this case gave rise to serious free speech concerns. The information to which it related was highly relevant to a series of policy decisions being taken by Government about how to safeguard those whose safety had been jeopardised by the release of the dataset. The grant of a super-injunction had the effect of completely shutting down the ordinary mechanisms of accountability which operate in a democracy. This led to what I described as a “scrutiny vacuum”.
Nonetheless, I decided on 23 November 2023 that the super-injunction should be continued, based on the MOD’s assessment that there was a “real risk that (i) the Taliban do not already know about the compromise of the dataset; (ii) disclosure of the fact of the dataset would cause them to take steps which lead to their obtaining it; and (iii) in that case, many thousands whose details are included in the dataset could be killed or injured and the UK Government would have no realistic way of safeguarding them”: see [43]. At that stage, the aim was to allow time for a safeguarding plan to be formulated.
By the time of my Judgment no. 2, on 15 February 2024, a plan had been formulated, though some of the details were still under discussion. I decided that the injunction should be continued again. The central basis for that decision can be seen from [31(a) and (b)] of my judgment:
“(a) There is a real possibility that, had the injunction not been granted, hundreds of ARAP eligible persons who have left Afghanistan since 1 September would not have been able to do so and would instead have been targeted by the Taliban for extra-judicial killing or serious physical harm. There is a further group which the Government has already decided to relocate to the UK. These are the existing ARAP-eligible persons and their dependants currently in Afghanistan, together with the very small additional cohort of about 200 high profile individuals and their dependants, who have been identified for relocation in light of the data incident. This group comprises a little more than 4,000 people. There is a real possibility that continuation of the injunction will save the lives of these people.
(b) There is a further group, some of whom may be offered relocation to the UK. This group comprises the 2,000 Triples and others whose ARAP eligibility is to be reconsidered and the 9,000 additional family members whose eligibility may in due course be reconsidered. There is a real possibility that the continuance of the injunction will allow some of this group to leave Afghanistan when they would otherwise be unable to do so. If it does, it may save their lives.”
But there was real uncertainty about the effect of the super-injunction on a much larger group affected by the disclosure of the dataset, whom the Government had decided would not be relocated or otherwise assisted. It was very likely that the existence of the dataset would come to the Taliban’s attention “in a matter of months” and continuation of the super-injunction would mean that they could not be forewarned of the danger to them, or given compensation which would enable them to take action to avoid that danger: [31(c)]. The possibility that the continuation of the super-injunction was having an adverse effect on this cohort, and the adverse effects on public and Parliamentary scrutiny (see [31(d)-(f)]), made it important to obtain a clearer evidential picture. So, while continuing the super-injunction, I directed an evidential hearing at which a representative of the MOD would give oral evidence and face questioning by counsel for the media representatives in OPEN (but in private) and by the Special Advocates in CLOSED.
The evidential hearing took place on 30 April and 1 May 2024. By that time, there had been further policy decisions about the response to the disclosure of the dataset. It had been decided to offer access to relocation to a further cohort of 2,300 individuals. When immediate family members were included, this amounted to 11,500 people. The costs of relocating this number of people and of housing them in the UK would be “spread over years” and could be assumed to run into “several billion pounds: the sort of money which makes a material difference to Government spending plans and is normally the stuff of political debate”: see [24] of my judgment handed down on 21 May 2024 (“Judgment no. 3”, [2024] EWHC 1220 (KB)). This meant that the cohort who had been offered relocation to the UK amounted to about 20,000 people: see [34].
My approach in Judgment no. 3 was to consider the effects of the super-injunction on two cohorts: first, the cohort being offered relocation (see [34]-[43]) and secondly, the cohort not being offered relocation (see [44]-[45]), while also taking into account the effect of the super-injunction on public debate (see [46]-[47]). In reaching my conclusion I said at [48] that the use of intelligence assessments to support the super-injunction was unusual in three respects. I set these out as follows:
“49. First, the risk assessments are caveated and there are a number of imponderables. It is tempting to take each assessment as a building block upon which to make further assessments and use this process to generate an overall view as to what is likely to happen if the super-injunction is discharged. But there is a real danger that, in doing so, one builds an edifice with very unsure foundations, ignoring the consequences that may be ensuing or may yet ensue if some of the initial assessments are wrong.
50. Second, even on the MOD’s assessments, there is a significant risk that the relief granted to date has in fact put lives at risk and is still doing so. This risk must be factored into the balance of risks and benefits.
51. Third, the relief granted to date not only prevents public discussion of the full reasons for the Government’s policy. It prevents the public from knowing of the very existence of the policy or the problem which it addresses; and it deprives decision-makers of information, public and Parliamentary scrutiny, all of which are liable to improve the quality of their decisions.”
At [53], I recorded my view that, while the super-injunction was on balance likely to be having a protective effect on the relocation cohort, there was a significant chance that it was in fact endangering some of them. The effect of the super-injunction on the larger non-relocation cohort was likely to be adverse overall.
These conclusions were based on findings that, if the Taliban already had the dataset, they may currently be using it to target people whose data is included in it. If so, the continuation of the super-injunction might be harming people by depriving them of the opportunity to take protective action: see [38]. Even if the Taliban did not have the dataset, on the MOD’s own assessment, they were likely to acquire it in the next few months or years; and those who had not been relocated at that point would be better off learning of its existence from the MOD (in advance of the discharge of the injunction) than from a “knock on the door by the Taliban”: see [41]. Moreover, the sheer scale of the decision-making, in terms of the numbers involved and the financial cost, meant that further secrecy was not feasible and was objectionable in principle: see [46]-[47]. I therefore ordered that the injunction be discharged with effect from 21 days after the handing down of my judgment, but stayed my order pending appeal.
The MOD appealed to the Court of Appeal. That court decided ([2024] EWCA Civ 838, “CA Judgment”) that I had made two main errors. First, I had failed to compare what would happen if the injunction were discharged with what would happen if it were continued. “On the undisputed evidence,” they concluded, “the effect of discharging the injunction would be to turn what was a real possibility into a virtual certainty”: [61]. This was a reference to the Taliban acquiring the dataset. Secondly, there was an error in relation to the finding of an adverse effect on the non-relocation cohort. It was wrong to consider that public pressure might assist that cohort, because there would not be time for that pressure to bear fruit in the form of a change in policy: see CA Judgment, [59]-[65]. The Court of Appeal went on to conclude that there was no material change from the position in my Judgment nos 1 and 2 and that the super-injunction should be continued, subject to periodic review by me at least every three months, noting that “the usefulness of the dataset to hostile actors is likely to reduce over time”: see CA Judgment, [80]-[81].
The media defendants did not apply for permission to appeal to the Supreme Court.
Events since July 2024
Since the handing down of the Court of Appeal’s judgment I have kept the super-injunction under regular review as required by that court. In doing so, I have applied to the best of my understanding the approach mandated in its judgment. That approach focuses on the harm which, on the basis of the MOD’s assessments, discharge of the injunction would, on balance, be likely to cause to the relocation cohort.
The Court of Appeal made clear that the case for continuation of the super-injunction might be undermined by evidence which affected the assessment that discharge would cause such harm to that cohort. In my regular reviews of the super-injunction, I have therefore been required to look for evidence which might change that assessment in a material way, including (though not limited to) evidence about the diminishing value of the dataset over time.
I received regular evidential updates and held OPEN (private) and CLOSED hearings on 11 and 15 November 2024 and 20 and 21 February 2025. Although there were evidential developments which were germane to the assessment in various ways, there was nothing to indicate a sufficiently material change in the evidential position before the Court of Appeal to justify discharge of the injunction, applying the approach set out in their judgment.
Prior to the latter hearing I was told that the Government was about to undertake a review of the relocation policy, to be conducted by a retired civil servant who had not been involved in any of the decisions under review. It was made clear by the MOD that this review might provide the kind of material change that could affect the balance. The review was initially to be conducted in the spring of 2025 but took longer than anticipated. I gave directions that, as soon as the report had been prepared it was to be placed before the court.
There were further hearings on 9, 19 and 23 May 2025, in part to deal with an application to vary and clarify the injunction by a firm of solicitors, Barings Law, representing more than 600 claimants who were aware that there had been some kind of data breach and were considering bringing a data protection claim. In parallel, I gave further directions determining which of the material before the court could be published in the event that the injunction were discharged. A further CLOSED hearing on that issue was listed for 26 June 2025.
The discharge of the super-injunction
On the day before this hearing, 25 June 2025, I received a copy of the review report, prepared under the supervision of the retired civil servant Paul Rimmer. An OPEN version of that report is annexed to this judgment. I do not, therefore need to summarise it. It is sufficient to say that it includes the conclusion, with respect to individuals whose data is included in the dataset, that acquisition of the dataset by the Taliban is “unlikely to substantially change an individual’s existing exposure given the volume of data already available”. It also includes the conclusions that “it appears unlikely that merely being on the dataset would be grounds for targeting” and it is “therefore also unlikely that family members—immediate or more distant—will be targeted simply because the ‘Principal’ appears in the… dataset”.
These conclusions fundamentally undermine the evidential basis on which I (in my Judgment nos 1 and 2) and the Court of Appeal relied in deciding that the super-injunction should be continued. I made clear that this was my provisional view at a CLOSED hearing on 1 July 2025, at which the Special Advocates submitted that the super-injunction should be discharged. On 4 July 2025, the Government Legal Department confirmed that the Secretary of State had decided to discontinue the relocation programme and to apply to discharge the injunction.
At a CLOSED hearing on 7 July, I heard submissions from Special Advocates acting in the interests of individuals involved in judicial review proceedings against the Secretary of State for Defence. Since the winter of 2023-24, they had been made aware of the existence of the dataset and of the super-injunction. They sought to persuade me to modify the super-injunction to enable the OPEN representatives in their cases to make submissions opposing its discharge. I declined to do that, for reasons which I gave in CLOSED. An OPEN version of those reasons will be made available separately.
In essence, there is no plausible basis on which a challenge to the conclusions in the review report would have any real prospect of success, given the wide range of sources on which it is based and the respect which the courts are required to give in public law proceedings to predictive assessments of the kind it contains. In light of the conclusions of the review, there is no tenable basis for the continuation of the super-injunction. This is particularly so given the serious interference it involves with the rights of the media defendants to freedom of expression and the correlative right of the public to receive the information they wish to impart.
I directed that the super-injunction should be discharged at 12 noon on Tuesday 15 July 2025, to allow time for affected persons to be informed before the general public and for further protective measures (which cannot be described in OPEN) to be implemented. I had in mind that news of the underlying matters should be made public before the Parliamentary recess.
The application for a further contra mundum injunction
On Tuesday 8 July 2025, I heard an application by the MOD for a separate contra mundum injunction imposing limited further restrictions on the disclosure of the dataset, extracts from it or information derived from it. There were further OPEN (private) and CLOSED hearings in relation to this on 11 and 14 July 2025. At a private hearing on Monday 14 July 2025, I announced my decision in principle to grant an interim injunction in terms much narrower than those sought by the MOD, pending a further hearing.
The precise terms of the interim injunction can be seen from the order, which is made available alongside this judgment. In essence, it will permit full reporting of almost all the relevant circumstances surrounding the data incident, but will prohibit anyone who has, has had or comes into possession of the dataset or information derived from it from: (i) disclosing the personal data contained in it; or (ii) describing the types of information contained in the “case notes” column. It will also require those in possession of the dataset or extracts of it to deliver up or destroy their copies if requested to do so. Accredited journalists and legal professionals will be exempt from this requirement. The parties will have a further opportunity to file OPEN and CLOSED evidence relevant to the continuation of the interim injunction. There will be a further hearing on Tuesday 22 July 2025 to consider whether to continue it and if so in what form.
Matters arising from the review
The contents of the report of the review raise questions, some of which require further investigation by the court. There is a CLOSED matter about which I have asked a series of questions and have asked for further submissions. It would not be appropriate to say anything further about it at this stage.
There is also another matter. The assessments in Mr Rimmer’s report are very different from those on which the super-injunction was sought and granted. The change is in part due to the passage of time. However, it also reflects information gathered from within Government, from speaking to third parties with knowledge of the situation in Afghanistan and from considering open-source documents. I have recently received a CLOSED witness statement which explains which of this material was available to those who undertook the initial assessments. The MOD has indicated that it intends to disclose an OPEN version of this statement today. It will be for others to consider whether lessons can be learned from the way the initial assessments in this case were prepared and whether the courts were, or are generally, right to accord such weight to assessments of this kind.
Postscript
Those involved in this long-running and unprecedented case have known throughout that there would come a time when the super-injunction could no longer be maintained. I decided that this point had been reached over a year ago. The Court of Appeal disagreed. For the last year, my assumption has been that the injunction might fall to be discharged when the information protected by it leaked into the public domain through the media in the UK or abroad. The parties have updated the court on a continual basis about the extent to which knowledge of the underlying matters has spread. It is one of the many remarkable features of the litigation—and very much to the credit of the media organisations and individual journalists involved—that there has been no mention in the media of the underlying matters while the super-injunction remained in force.
ANNEX
This gist of the Policy Context Review has been prepared from a fuller CLOSED version. Where material has been redacted from the statement to allow for it to be disclosed in PRIVATE, this is not indicated on the face of the document, save where necessary in the interests of clarity. In these circumstances, a gist is shown in italics within square brackets.
POLICY CONTEXT REVIEW
PAUL RIMMER, JUNE 2025
INTRODUCTION & NOTE ON SOURCES
1. This is an independent review of the policy context for the data breach, now that 3 years have passed since the unauthorised disclosure of Afghan Relocations and Assistance Policy (ARAP) applicants’ data in February 2022. This review considered Afghanistan in Spring 2025, and reviewed HMG’s proposed policy response against this.
2. This review is a risk based review, which seeks to examine the overall policy context. The focus of this review is on the potential impact of the dataset itself, and specifically how this might increase the net additional risk to individuals affected.
3. Part I considers whether HMG’s understanding of the level of risk faced by those affected by the data incident remains broadly current, by building on our insight into the “lived experience” of those in Afghanistan and the Taleban’s current activity. Part II provides some reflections on the current response.
4. The review has been conducted by Paul Rimmer, a retired civil servant, supported by a small team of Cabinet Office civil servants. The reviewer has a background in intelligence and policy matters and has not previously worked on the current Afghan Response Route (ARR) policy.
5. This policy context review draws on existing work, including taking account of existing assessments, expertise and reflections from current Afghanistan work from across HMG, Foreign, Commonwealth and Development Office (FCDO) Posts across the region, as well as external stakeholders. The review focused on those most able to provide a high level of insight into the current situation as of Spring 2025, including:
a. Open source reports from a wide range of NGOs and think tanks; and
b. Interviews with representatives from:
i. An NGO;
ii. UNAMA;
iii. Afghan cultural experts and academics;
iv. Afghans already resettled in the UK;
v. Afghan and international activists working on Afghan resettlement;
vi. International allies.
6. Written reports and evidence, both publicly available and internal reporting, were considered, but the majority of this review has been conducted through in-person interviews. Some of these individuals are aware of the incident and therefore able to speak directly to the nature of the risk posed by the dataset specifically. A number of these have previously provided information through the Court process for HMG to assess, some of which was re-examined as part of this review for thoroughness. Others provided expertise more generally, on the Taleban approach to [certain individuals].
7. This review has also examined the dataset itself, in order to consider its contents against the insights provided by the various interlocutors interviewed. This was not a detailed analysis of each data line, but rather to gain a general understanding of the type of data included.
8. This review relies on open source, and often anecdotal evidence. This approach has been adopted deliberately in order to make use of information sources that are current, but inevitably difficult to assess individually. Nevertheless, they combine to give a broadly complementary picture. We sought to thoroughly scrutinise the credibility of the evidence provided from all sources drawn on. In examining the information we have been provided, the motivation and expertise of the source was considered, along with how their information was obtained and its currency. Where possible, corroboration with other sources has been sought, drawing out themes common across multiple interlocutors.
SUMMARY
9. The challenge for HMG is to design a response to the data breach in a manner that is proportionate to the risks that the unauthorised disclosure poses. This review concludes that, nearly four years into the de facto Taleban rule of Afghanistan, early concern about the extent of Taleban intent to target [certain individuals] has diminished. There is little evidence of intent by the Taleban to conduct a campaign of retribution against [certain individuals]. This review has also found no evidence to determine there is a consistent or sustained effort to target [certain individuals]. Killings are undoubtedly still occurring, and human rights violations remain extensive, but it is extremely difficult to determine the causes of individual killings or detentions.
10. The actual data set (versus some knowledge of “data” being lost) may not have spread nearly as widely as initially feared. Given the data they already have access to as the de facto government, we believe it is unlikely the dataset would be the single, or definitive, piece of information enabling or prompting the Taleban to act.
11. Vulnerability in Afghanistan is multi-layered. The severe and worsening economic deterioration, threat from the Islamic State Khorasan Province (ISKP), repressive police state and ultra-conservative interpretation of Sharia law combine to form a dire situation of deepening vulnerability for many across Afghanistan. In some areas, links to the former government may exacerbate a person’s vulnerability, but these links are likely to be already well known locally. Whilst it is impossible to determine definitively how useful the dataset would be to the Taleban, this review concludes, in the absence of evidence the Taleban are actively acting on it, that it is unlikely to substantially change an individual’s existing exposure given the volume of data already available.
12. Therefore, this policy review suggests that the ARR – as currently proposed – addresses the risks from the data incident in a manner that may now be disproportionate to the actual impact of the data loss were it fall into the Taleban’s hands.
KEY CONCLUSIONS
a. Appalling human rights abuses occur – including extra-judicial killings – against former officials. But there is also limited evidence to suggest that [certain individuals] have been targeted with any degree of consistency.
b. Given the nearly four years since the Taleban takeover, posing a current threat or resistance to Taleban rule is likely to be a far more persuasive factor in the threat faced by individuals in Afghanistan, rather than former affiliations. As such, it appears unlikely that merely being on the dataset would be grounds for targeting. It is therefore also unlikely that family members – immediate or more distant – will be targeted simply because the “Principal” appears in the dataset.
c. Should the Taleban wish to target individuals the wealth of data inherited from the former government would already enable them to do so. Additional data is always of interest to develop leads for investigative or targeting purposes. Publicity about the dataset’s loss would inevitably raise interest in acquiring it. But it is highly unlikely it would be the single, or definitive, piece of information enabling or prompting the Taleban to act. It is a “piece of a puzzle” rather than a “smoking gun”.
d. No evidence points clearly to Taleban possession of the dataset. It is plausible the dataset (its actual content, rather than knowledge that a dataset exists) has not spread as widely or as rapidly as was initially feared.
e. Given this context, the current ARR policy appears an extremely significant intervention, with not inconsiderable risk to HMG and the UK, to address the potentially limited net additional risk the incident likely presents. Based on the conclusions of this policy review, and the level of risk inherent within the current ARR policy, HMG could consider amending the approach to reflect the value the dataset offers. The Taleban already have access to significant volumes of data which enables them to identify personnel associated with the former government. The family and community based nature of Afghan society means former roles and associations are often already well known. The dataset is unlikely to significantly shift Taleban understanding of individuals who may be of interest to them. As a result, it is unlikely to profoundly change the existing risk profile of individuals named on the dataset.
PART I: THE CONTEXT IN SPRING 2025
DATA BREACH
13. In February 2022, an unauthorised disclosure from the Ministry of Defence (MOD) compromised 33,000 lines of personal data that links to c.18,800 principal ARAP applicants and details of their family members.1 Of these c.18,800 applicants, the majority (over 15,700) had been found ineligible for ARAP. MOD became aware of this incident on 14 August 2023 when an anonymous user exposed via Facebook the names and personal details of ten individuals from the wider dataset.
DEFENCE INTELLIGENCE ASSESSMENT
14. A Defence Intelligence (DI) assessment was originally undertaken in [date], and last formally reviewed in September 2024, to assess the threat posed to the individuals.
1 While all affected individuals applied for ARAP, some also applied for ACRS.
15. DI recognises the value of this review and its conclusions, reflecting the developing dynamics in Afghanistan and the fact of this being a risk-based review of the overall policy.
RELOCATIONS
16. To date (as at May 2025), HMG has relocated 35,245 individuals to the UK from Afghanistan and other countries under ARAP, ARR and ACRS. Of this number, 16,156 individuals affected by the data breach have already been relocated to the UK.
THE TALEBAN OF 2025
17. The Taleban has been the de facto government of Afghanistan since August 2021. In practice, the Taleban is a group of factions made up of different tribes, ethnicities, backgrounds and histories. This, and a lack of central, written policies, mean that implementation of senior leadership intent can vary, although the Taleban as a whole exert effective control across the entire country. Kandahar remains the Taleban’s centre of power, with Emir Hibatullah Akhundzada exercising power from this part of southern Afghanistan. He, and a small group around him, rule through issuing edicts to the wider Taleban.
18. Regime security remains a priority for the Taleban. This Kandahar-based leadership is increasingly at odds with the Taleban ministers in Kabul, who are focused on daily governance of the ministries. The suspension of US Aid (more below) may negatively impact certain ministries and their remit.
19. The Taleban’s current priorities are [details given]. Afghans who have resettled in the UK report a significant increase in Taleban house searches across Kabul in 2025, as the Taleban seek to hunt out evidence of those actively resisting their regime.
20. Recent developments have highlighted the Emir’s intent to enforce his rule over the Taleban and Afghanistan. For example, in February 2025, the Emir deployed Kandahar-based troops to Kabul, to control Kabul airport and checkpoints connecting the capital city to other provinces. It is likely this was in response to the departure from Afghanistan of Talib Deputy Minister of Foreign Affairs, Sher Mohammed Abbas Stanikzai, who had openly criticised the ban on girls’ education.
ECONOMIC DETERIORATION
21. In 2024, an estimated 23.7 million Afghans, more than half the population, required humanitarian assistance. Despite the scale of the need, as at May 2024, it was estimated that only 28% of those requiring humanitarian assistance were receiving it. UN and NGO costs are funded by UN shipments of c.$40m per week, which indirectly stabilise the currency. To date, the US has been the principal aid contributor, providing 47% of Afghanistan’s humanitarian funding in 2024. President Trump suspended foreign aid in January 2025, including to Afghanistan. Given its reliance on US assistance, Afghanistan faces one of the most severe impacts from the US funding suspension globally. The suspension also threatens global vertical funds – including GAVI, the Global Fund, GEF and GCF, which play a critical role in sustaining health and climate programming in Afghanistan.
22. What this means in this context is likewise not clear cut. But anything which exacerbates the existing levels of instability across Afghanistan is likely to fuel violence and potentially threatening coping strategies. On the one hand, increased civil unrest may see greater focus on regime and individual survival, with less focus on ‘retribution’ against those linked to the former Government or international forces. On the other hand, as families find their economic situation ever more precarious, they may seek more opportunities to exploit vulnerabilities where they can find them.
HUMAN RIGHTS LANDSCAPE
23. Whilst the Taleban of 2025 cannot be automatically viewed as the same organisation that controlled Afghanistan between 1996 and 2001, it remains a theocratic police state intent on implementing its highly conservative version of Islam across Afghanistan. Human rights organisations and Afghanistan watchers have documented egregious human rights violations and many groups in Afghanistan continue to face a real risk of persecution.2
24. Compared to the 1990s, the de facto Taleban regime seems more aware of the need to control the narrative, especially internationally, and are conscious of what evidence is available of its human rights violations. Anecdotal reports suggest mobile phones have been removed from individuals before beatings and other abuses, before being returned, to ensure no video/photographic evidence can be obtained. The media industry, which had flourished under the former Government, has been effectively crushed by the Taleban authorities, further limiting the detail around human rights abuses and killings that are emerging from Afghanistan.
25. Regardless, it is clear that the human rights picture is dire. Repression and human rights violations are widespread across Afghanistan. Levels of violence, although reduced since the end of the conflict, remain high. The Taleban’s ever expanding erosion of women’s and girls’ rights is their defining policy, and repression of civil society and discrimination against minority communities is well documented publicly. The Special Rapporteur on Afghanistan, Richard Bennett, has documented consistent reports of torture, ill treatment, and enforced disappearances, by Taleban officials, in particular by the de facto General Directorate of Intelligence and de facto police.3
2 See for example: Human Rights Watch Work Report 2024; Report of the Special Rapporteur on the situation of Human Rights in Afghanistan (30 August 2024); Amnesty International Report 2023/2024: Afghanistan; Centre for Information Resilience: Afghan Witness.
THE TALEBAN GENERAL AMNESTY
26. Considering the human rights landscape in the context of the data incident, this review has sought to determine, based on several years of de facto Taleban rule, whether [certain individuals] remain as likely to be specifically targeted, alongside or ahead of, minority and marginalised groups.
27. After taking control of Afghanistan in August 2021, the Taleban announced a “general amnesty” for former officials of the Islamic Republic of Afghanistan and former members of the Afghan National Defence and Security Forces (ANDSF). No publicly released written guidance has been issued by the Taleban, but senior members of the Taleban authorities have verbally reiterated their commitment to the amnesty in public, and through diplomatic channels to the FCDO. This ‘amnesty’ has its roots in a pre-takeover Taleban policy. In 2016 the Taleban confirmed an official policy of offering amnesty to those associated with the former government or international forces, if they returned to civilian life, surrendered their arms, apologised for harm done and had a trusted elder to vouch for them.4
28. However, reports abound of the Taleban not respecting this amnesty in practice, especially in the months immediately after their takeover. Most comprehensively, UNAMA issued a report monitoring the Taleban’s adherence to this amnesty in 2023, covering the period between August 2021 and June 2023.5 To date, this is the most detailed report published on the Taleban’s approach to former government officials and security members. Throughout this period UNAMA recorded 800 instances of human rights violations, including extrajudicial killings, arbitrary arrests, torture and ill treatment, carried out by the Taleban authorities. More recent (May 2024) UNAMA reporting records arbitrary arrests, torture and killings of former government officials and ANDSF members, although it does not provide a breakdown of the data based on victims’ profiles.6
29. Similarly, Human Rights Watch (HRW) published a report in November 2021, based on 67 witness interviews, documenting the summary executions or enforced disappearance of over 100 former members of the former Afghan security forces between 15 August and 31 October 2021, in Ghanzi, Helmand, Kandahar and Kunduz provinces.7 HRW also found evidence of targeting of family members, in an effort to access the principals.
3 Report of the Rapporteur on the situation of Human Rights in Afghanistan (30 August 2024).
4 A. Jackson, ‘Negotiating Survival: Civilian-Insurgent Relations in Afghanistan’, (C Hurst &Co: 2021), p. 95.
5 UNAMA, ‘A Barrier to Security Peace: Human Rights Violations against Former Government Officials and Former Armed Forces Members in Afghanistan’. August 2023.
6 UNAMA, ‘Update on the Human Rights Situation in Afghanistan: January-March 2024’, May 2024.
30. Much of the detailed information around human rights violations against this group relates to the period immediately post-Taleban takeover, with a reduction in numbers of killings reported since then. More recently, detailed surveys have mostly been conducted by Rawadari – an Afghan Human Rights NGO. Its Annual Human Rights Report for 2024 noted that the human rights situation in Afghanistan worsened across multiple factors in 2024.8 Rawadari’s findings indicate that former government employees and their families continue to be subjected to arbitrary detention, torture and killings. The rate of extrajudicial killings against this cohort increased by 9% in 2024 (from 83 to 91 documented cases). Rawadari notes that given the severe limitations imposed by the Taleban, it is extremely challenging to determine the causes for these killings and arrests. Some appear to be a result of accusations of collaboration with opposition groups, others reflect local grievances, and some acts of retaliation. Even so, available evidence suggests that absolute numbers of detentions and killings are much lower than in the immediate aftermath of the Taleban takeover in 2021.
TARGETING OR LOCAL GRIEVANCES?
31. Despite these continuing clear violations and human rights abuses, nearly four years on from the Taleban takeover the overwhelming consensus is that there is no longer a widespread campaign targeting [certain individuals] as many Afghan watchers had feared. This review specifically sought to develop current understanding of Taleban targeting given how critical this is to HMG’s policy response to date in ensuring that individuals at highest risk are relocated. This review agrees that were the Taleban to seek to target [certain individuals] they would be likely to prioritise high profile individuals; such as the Triples.
32. This review concludes that there is little evidence of targeting of such individuals in Spring 2025, and in particular, deliberate targeting is unlikely to be solely on the basis of an individual appearing on the dataset. Additionally, this review assesses that the threat to [certain individuals] has probably lowered since the first few months of the Taleban takeover, as the de facto regime is predominantly focused on targeting those perceived to threaten their rule now (more below). Nonetheless, it is clear that life in Afghanistan for those who perceive a threat from the Taleban is miserable, regularly moving location and relying on family support to hide. There is a general atmosphere of anxiety across Afghanistan, with [certain individuals] living with the fear that the Taleban could decide to persecute them.
33. Retribution against [certain individuals] does not feature heavily in UNAMA’s regular human rights reporting.
7“No Forgiveness for People like You”, Human Rights Watch, 30 November 2021.
8 Rawadari, ‘Afghanistan Human Rights Situation Report 2024’, March 2025.
34. In response to the UNAMA report, the Taleban asserted that any breaches of the amnesty and human rights abuses were based on “personal enmity or revenge”, and not an official order. Whilst this statement clearly cannot be taken at face value, given the decentralised nature of Taleban governance and the embedded nature of local disputes and grievances, it is highly likely that this is a factor in some targeting of individuals linked to the former Afghan government and international forces. The European Union Agency for Asylum Country Focus report for Afghanistan, covering the period from 1 July 2022 to 30 September 2023 found the approach to former officials to be inconsistent, and that targeting appeared driven by local contexts rather than an individual’s association with the former government.9
35. As such, although those on the dataset may be targeted, not all incidents could be concluded to be evidence of a campaign of persecution against those associated with NATO or the Afghan security apparatus, or even that they were opportunistically targeted because they are in the dataset. It is likely many are also a result of involvement in anti-Taleban activities, personal disputes, rivalries with individual Taleban members, or reflective of wider Taleban persecution of minority groups.
36. It is likely “score settling” or retribution attacks continue in local communities, including being perpetrated by Taleban officials. Nearly four years have now passed since the Taleban takeover, there was consensus that there is limited evidence that this is a widespread phenomenon amongst Afghan communities, in the way that was witnessed in the immediate months after the Taleban came to power.
37. Evidence from multiple interlocutors indicates it is increasingly likely that those who the Taleban are now focused on persecuting – nearly four years into their de facto rule of Afghanistan – are those who have the capability or skills to effectively resist or undermine Taleban rule.10 The active targeting of individuals by the Taleban authorities appears more likely to be driven by assessments on who poses a threat to them now.
TALEBAN POSSESION OF THE DATASET
38. The Taleban are highly likely to be interested in any data available. It remains likely they would wish to get hold of the dataset, if they do not already have it, given the unprecedented efforts HMG have gone to protect it. Publicity around the dataset would inevitably raise interest in acquiring it, although it is not clear that this would be sufficient to alter the Taleban’s approach to targeting [certain individuals] given the de facto regime’s priority appears to be [evaluation given].
9 EUAA, Afghanistan – Country Focus. December 2023, p. 89.
10 Given the focus of this report, wider human rights abuses have not been covered. Taleban targeting of minority and marginalised groups (e.g. based on ethnicity or sexuality), and continued widespread and egregious human rights violations against women and girls are widely documented in wider reporting.
39. It remains the case that it will never be possible to determine with absolute certainty that the breach has been ‘contained’.
40. The multiple reports of discussion amongst Afghan communities in the UK and elsewhere, about the existence of “data”, seem to be without knowledge of what this comprises and it is hard to determine whether this in fact refers to this, or other datasets, given how much is out there.
41. Recent developments relating to the private law claims Barings are proposing to bring on behalf of their clients (“the Barings claims”), do not at present change this judgement. The numbers referenced by Barings indicate knowledge of a data breach, including within Afghanistan, but not necessarily that the dataset itself is widely circulating. Barings have confirmed they have 667 clients: 504 in the UK; 154 in Afghanistan; 3 in the US; 3 in Turkey and 3 in other countries. Barings clients are individuals who believe they are affected by the unauthorised data breach, so it may be reasonable to assume that they therefore are likely to have an incentive not to circulate it further if they do have access. However, if knowledge of the data breach is more widely circulating, this may increase Taleban interest in obtaining it to understand whether it is of value to them.
42. Whilst there are many uncertainties, in how the Taleban will respond, if further evidence were confirmed, by Barings or others, that the dataset was already widely available, including to the Taleban, this may, in part, support the emerging picture set out in this review. To date, HMG has been gravely concerned about how the Taleban would respond against individuals affected. The current understanding of the Barings claims does not at present alter the judgments reached by this review although HMG will need to keep this under close review.
DATASET “EXPLOITABILITY”
43. After the Taleban takeover, Afghan military units sought to destroy IT and burn documents to prevent operational information becoming available to the Taleban. However, all payroll, pensions and service details for military and police employees were held by the Ministry of Interior and the Ministry of Defence, and were handed over to the Taleban. All ANDSF personnel were paid via Kabul Bank and required to be treated in military hospitals – all of these medical records and financial records were transferred to the Taleban. Personal details of officials working for other Afghan government departments, were likewise transferred to the Taleban. Human Rights Watch reporting confirmed that when taking over the National Directorate of Security, the Taleban were able to obtain not only data on employees, but also information on those who might have acted as informants.11 Afghans who have resettled in the UK also shared evidence of the Taleban using social media as a primary method for identifying those they are interested in who [held certain roles].
11 “No Forgiveness for People like You”, Human Rights Watch, 30 November 2021.
44. Most critically, the biometric details held by the former government are all believed to have transferred to the Taleban in August 2021. The Ministry of Interior’s biometric database, the Afghan Automatic Biometric Identical System (AABIS/Biometrics Center), was used to check criminal and terrorist backgrounds of all applicants for government jobs, and for passport, national ID, driver’s licence applications, and registrations for Afghanistan’s college entrance exam.12 Likewise, the US-funded Afghan Personnel and Pay System (APPS) was used by both the Afghan Ministry of Interior and Ministry of Defence to pay the national army and policy. Estimated to hold some half a million records, it held extensive detail about security personnel and their extended networks. Each profile in APPS is understood to hold at least 40 data fields.13 This includes names, date and place of birth, as well as the unique ID number that connects each person to the biometric profile held by the Ministry of Interior. Family and “tribal elder” connections are mapped out, with the individuals career trajectory. Evidence of the Taleban exploiting biometric data predates their takeover. In 2016 a widely reported incident saw insurgents hijack a bus and use a fingerprint scanner to identify Afghan National Army soldiers.14 Additionally the Afghan National Biometric System used to issue Afghan national identity cards (e-Tazkira), and the National Directorate of Security and the Afghan Supreme Court payrolls were underpinned by biometric databases. A number of interlocutors suggested to this review that with this wealth of biometric data, the Taleban may place less value on other datasets.
45. Human Rights Watch documented evidence in their November 2021 report, of the Taleban making use of access to employment records left by the former government, to identify ANDSF and former security personnel for arrest and execution between August and October 2021.15
46. This review examined the dataset itself, and the information contained in it. The level of detail varies considerably between each data line. Whilst some entries contain more detail, others barely provide enough information to identify an individual. It is therefore extremely challenging to determine level of risk caused by the dataset as a whole, or even in many cases, for specific individuals.
47. Whilst it is impossible to determine definitively how useful the dataset would be to the Taleban, this review concludes that given the extent of the data already available, the dataset is unlikely to provide considerably new or highly pertinent
12 MIT Technology Review, ‘This is the real story of the Afghan biometric databases abandoned to the Taliban’, August 2021.
13 MIT Technology Review, ‘This is the real story of the Afghan biometric databases abandoned to the Taliban’, August 2021.
14 Tolo News, ‘Taliban Insurgents behind the recent mass kidnapping in Kunduz, used a government biometric systems’, June 2016.
15 “No Forgiveness for People like You”, Human Rights Watch, 30 November 2021.
information to the Taleban. With Taleban targeting increasingly focused on those currently posing a threat to their rule, the dataset is therefore extremely unlikely to offer significant additional information of use to the Taleban in determining this. As such, it appears highly unlikely that merely being on the dataset would be grounds for targeting. It is therefore even more unlikely that family members – immediate or more distant – are likely to be targeted simply as the principal appears in the dataset.
48. In addition, given the tribal and interconnected familial nature of Afghan communities, it is highly likely that any connections to the former government, security services or international forces were well known both before, and in the years since, the Taleban takeover. It is therefore extremely unlikely that the dataset would provide the primary, or first, confirmation of associations with the former government or other nations. Similarly, it is likely that local communities will know that those on the dataset who had falsely claimed to have held specific roles, in order to seek eligibility for relocation, did not in fact have those connections.
OVERALL RISK CONTEXT FOR THE DATASET
49. Given the available evidence and the widespread human rights violations, it is not possible to conclude decisively that those individuals affected by the data incident would not face any additional threat, if the dataset were acquired by the Taleban or more widely known publicly in Afghan communities.
50. The glare of publicity around revelation of the data loss would clearly be likely to attract Taleban interest in obtaining it. It is possible that HMG has inadvertently added more value to the dataset by establishing a bespoke scheme, and through the use of an unprecedented super-injunction. Given the cost and scale of the proposed ARR pipeline, as this becomes public, it may perpetuate a perception that the dataset provides information of considerably higher value than this review judges it to in reality. Public knowledge of the data breach could also be of interest, for example, to hostile state actors seeking avenues to embarrass or criticise the UK.
51. Therefore, rather than being the defining factor in an individual being targeted, it seems increasingly likely that public knowledge of the dataset would be simply another factor in exacerbating a person’s existing vulnerability. As such, it is difficult to argue knowledge of the dataset will automatically mean those named are immediately at substantially greater risk. In circumstances in which the dataset enables a local community to definitively confirm a link to international forces, this may put an individual at higher risk. However, given the tribal and interconnected nature of Afghan communities, and the extent of the data already available to the de facto Taleban authorities, it is highly likely any such connections will already be known, regardless of public knowledge of the dataset because of the continued lack of attention being paid to this community by the Taleban.
52. Media reporting has detailed confirmed incidents in which extended family members of “high value” individuals have been targeted, including arrests, detentions and torture. However, given the more limited risk the data incident is now judged to present to Principals named in the dataset (excluding those judged high risk by other resettlement schemes), it follows that the risk to family members and more distant associates is also correspondingly lower.
PART II: REFLECTIONS ON THE POLICY RESPONSE
53. The review notes that the passage of time is particularly relevant.
54. In conversations conducted throughout this review, a number of factors beyond the risk posed to affected individuals were repeatedly raised, that HMG may need to consider in taking any decisions on the ARR. These principally related to issues around the consequences for scrutiny and transparency, given the unprecedented legal action it has required, and corresponding discussions around value for money. Likewise, many noted these pressures are particularly challenging given the UK domestic housing system is under acute pressure; with record levels of homelessness and wider public service pressures including from supporting existing Afghan cohorts, which are particularly acute in certain areas, and risks to community cohesion. These are all key areas for public debate, with parliamentarians and the media expecting to provide scrutiny to HMG’s approach in these areas. The super-injunction that has been granted to cover the policy response to the data incident has stopped this scrutiny being possible.
55. This review focused on the risk posed to individuals affected by the dataset, rather than broader risks to HMG. However, it is worth reflecting that several interlocutors expressed concern around the risk that resettled Afghans could be radicalised in the UK. There is a risk of a growing gap between resettled Afghans’ expectations, and the reality of what ever-more stretched domestic services can deliver. Some interlocutors also highlighted concerns around the extent to which Afghanistan is becoming a base for a wide range of terrorist groups. This review strongly recommends that this should be an area for further work, and the Home Office will want to consider this in more detail.
56. Based on the conclusions of this context review, balanced against the wider implications to HMG from the ARR, HMG could consider amending its approach to reflect the value the dataset offers. To date, the response has taken a cautious approach to risk. HMG planning has been based on the underpinning conservative risk judgement that were the Taleban to secure access to the dataset, the consequences for affected individuals may be serious. As a result, it was felt that this was appropriate. However, this review of the context into which the dataset would be released within Afghanistan suggests that the additional threat affected individuals would face might be more limited. Rather than being the defining factor in an individual being targeted, it is likely that public knowledge of the dataset would be simply another factor in exacerbating a person’s existing vulnerability. As such, it is difficult to argue Taleban knowledge of the dataset would automatically mean those named are immediately at substantially greater risk. As a result, the current ARR policy appears a significant intervention, against potentially limited additional net risk.
RECOMMENDATION
57. Consideration should be given to whether there are other avenues to seek insights from Afghans who are resettling in the UK. This could be conducted as part of the reception and immigration process.
ANNEXES
Defence Intelligence Threat Assessment
Source List
Letter from CDI to SRO ARU
Like Minded Countries – Lived Experiences
SOURCE LIST
Many of the external individuals spoken to directly asked not to be named in relation to this review.
HMG Internal
• FCDO
◦ Andy McCoubrey – FCDO Director
◦ Afghan Policy Team
◦ Afghan Resettlement Team
◦ Research Analysts
◦ UKMis – Robert Chatterton Dickson
◦ BHC Pakistan – Jane Marriott
• Home Office
◦ Migration and Borders Group
◦ Aviation and Migration Policy (Homeland Security Group)
◦ Asylum Team
◦ Forced Returns Team
• Ministry of Defence
◦ Maj Gen Dan Blanchford – Op LAZURITE
◦ Natalie Moore – Dir DARR
◦ Defence Intelligence
External
• Person C
• Person A
• Afghans already resettled in the UK (from the Triples community)
• NGO
Main Open Source Written Material
• Human Rights Watch Work Report 2024
• Report of the Special Rapporteur on the situation of Human Rights in Afghanistan (30 August 2024)
• Amnesty International Report 2023/2024
• Afghanistan; Centre for Information Resilience
• Afghan Witness
• A. Jackson, ‘Negotiating Survival: Civilian-Insurgent Relations in Afghanistan’, (C Hurst &Co: 2021).
• Rawadari
• EUAA, Afghanistan – Country Focus, 2023.
• MIT Technology Review, ‘This is the real story of the Afghan biometric databases abandoned to the Taliban’, August 2021.
• Tolo News
Like-minded Countries
1. HMG routinely conducts engagement in Islamabad with like-minded countries operating similar resettlement schemes.
2. To date we have received no evidence of any organised targeting of like-minded countries’ eligible persons (EPs) as they leave Afghanistan through this forum.
3. There has been a significant number of Afghans sponsored and relocated by third countries (over one hundred thousand) which, when taken in aggregate, suggests that the Taliban are more than aware of movements.
4. The below information reflects the current known scale of support offered to Afghan EPs by like-minded countries to the UK.
5. Many countries have scaled down the level of support offered.