New IARMJ guidelines offer practical framework for social media evidence in asylum appeals

iarmj
The International Association of Refugee and Migration Judges

Established at Warsaw 18 September 1997

Judicial Guidelines for the Analysis of Social Media Evidence in Refugee, Protection and Migration appeals

Judges may wish to consider taking the following factors into account (cross-reference are to the relevant passages of the explanatory note below).

1. Reliability

• How easily could the account have been altered by the migrant/international protection applicant relying on it such that what is now provided at the hearing does not reflect the true underlying position over time? (§22-25)

2. Selectivity

• Does the evidence fairly reflect the full context of the asylum seeker's social media profile? (§21(e), 23)

3. Durability

• How long will it last/how long was it live online? (§10, 20-21)

4. Visibility

• Public/private settings, friends, "shares", "likes", links from other platforms (§4, 7, 9-11, 16-18, 20-21)

5. Deletability

• Could it be deleted and what would the fundamental rights consequences of that be? (§10, 14-15, 19-20, 21(f))

6. State enquiries

• Country of origin's authorities' interest in social media: monitoring expatriate activities, reaching out as "false friends", making enquiries on return (§9, 14, 16-18, 20-21)

7. Accessibility

• How can a relevant social media account be accessed safely and lawfully? (§7-8, 22-24)

8. Preservation of evidence

• How can relevant social media evidence be preserved and archived? (§21-25)

These Guidelines are published by the International Association of Refugee and Migration Judges on 1 October 2025. Copyright is asserted.

Explanatory note for the Guidelines on Analysing Social Media Evidence

From the IARMJ's Country of Origin Information, Expert Evidence and Social Media Working Group

(Rev1)

Introduction and definitions

1. The rise of social media has been one of the most significant phenomena of the 21st century. It is regularly encountered in every area of life where information technology is available. We hope it will be useful to take time in reflecting how members of courts and tribunals adjudicating in international protection and migration cases [1] should weigh social media in the balance alongside other evidence. [2]

2. By social media we refer to interactive technologies that facilitate the creation and sharing of information, ideas, interests, and other forms of expression through virtual communities and networks. Here is one definition propounded by a judge (speaking extra-judicially): [3]

"Social media refers to a variety of online platforms which are centred on social interaction. These platforms defy the traditional one-way model of distribution and consumption in other forms of media. In traditional models, such as print, TV and radio, content is created at a central source and distributed to consumers in a one-way, usually dead-end direction. Letters to the editor and talk back radio are limited exceptions within this traditional model. With social media, content is not merely consumed by users, it is also created, organised and distributed by them. Social media platforms thereby create a dialogue between different people, allowing them to communicate and share information."

3. Social media is an umbrella term that describes a vast variety of different types of interactive communication such as social network sites or social networking sites; blogs; wikis; podcasts; forums; content communities; microblogging; and messaging. [4] Social media applications (commonly termed "Apps") typically feature user-generated content (such as text posts by the author and comments by other users on the author's posts, though also digital photos and videos) and data generated through online interactions. Users create profiles specific to the service in question that are then maintained by the application. Active use of a particular application helps the development of online social networks by connecting a user's profile with those of other individuals or groups.

4. One form of social media is the blog (from "web-log") which is an informational website published on the World Wide Web consisting of discrete, often informal diary-style text entries ("posts", a term also given to any written contribution by a user via their account). [5] Typically posts are the subject of "likes" where approved by other users who may well distribute them onwards by sharing them with their own online circle of acquaintance.

5. There are numerous forms of social media, amongst the most commonly encountered are: Facebook, YouTube, WhatsApp, Instagram, TikTok, UTube and X (formerly Twitter). Around the world there are many others and different countries may restrict access to platforms deemed as against their cultural, religious, commercial interests or inclinations or against their real or perceived public interests: for example in China only LinkedIn is available amongst the more common internationally popular applications, [6] whereas in Iran, Instagram and Pinterest amounted to over 55% of all social media usage and Facebook only 15%. [7]

6. Not everyone engages in social media, of course, and the judicial community is doubtless much more reticent than some other professions. Nevertheless as noted by the Global Judicial Integrity Network: [8]

"Irrespective of whether they use social media or not, judges should have a general knowledge of social media, including how it may generate evidence in cases that judges may decide. Judges should also have an understanding of existing online communication tools and technology, including artificial-intelligence-powered technology."

Screening and investigatory powers of national authorities

7. National determining authorities take an increasing interest in social media. [9] Measures such as the EU's Screening Regulation [10] (the objectives of which include the collection and sharing of relevant information with competent authorities and may include the verification of objects in the possession of third-country nationals, subject to respecting their human dignity) and the Procedures Regulation [11] provide a legal context for examining asylum seeking arrivals. In 2019 the USA's Department of Homeland Security proposed collecting usernames from nineteen social networking sites. [12] Then government agents may access social media data (on the asylum seeker and their connections) not only by looking at publicly available data online but also by logging into social media platforms using constructed personas which cannot be traced back to the individual civil servant or to the institution, in accordance with internal guidelines: a practice intended to protect both civil servants and claimants. In Germany the analysis of digital devices to determine the nationality and identity of an asylum seeker or potential deportee is permitted without their consent [13] where necessary if no less intrusive measure is available; potentially this would authorise analysis of social media accounts (at least where stored locally on the device rather than in the cloud). [14]

8. Blanket policies to confiscate the mobile phones of asylum seekers need to be critically assessed as to whether they are in accordance with the law and the underlying statutory powers permitting such actions. [15] There may be national or regional norms which control the processing of data, such as the European Union's General Data Protection Regulation, which requires lawful, fair, transparent and secure accurate data collection procedures which retain only the minimum data which is essential to achieve a specific purpose. [16] UNHCR have warned of the need for safeguards against unwarranted seizure of devices: such measures should never become routine but should take place only where strictly necessary, regard should be had to lawyer/client privilege in reviewing the material thereby obtained, and it should be borne in mind that asylum seekers may well use false names to avoid endangering their families in the country of origin. UNHCR stress that regard should be had to the rights to human dignity, to private and family life, to protection of personal data, and to own, use, and dispose of one's lawfully acquired possessions. [17] In the era of artificial intelligence, there may be national or regional norms restricting the use of AI systems for remote biometric identification for law enforcement and similar purposes, such as the Artificial Intelligence Act in the European Union. [18]

Relevance of social media evidence to migration litigation

9. Social media evidence may be relevant in asylum determination in various ways, most obviously:

(a) In corroborating the truth of an account, because events asserted to have taken place are verifiably documented online.

(b) In controverting the truth of an account, because information comes to light which is inconsistent with the asylum seeker's story. [19]

(c) In determining the level of exposure a person's profile or views may have had, via the extent of their social network and via any express threats received. [20]

(d) In determining whether the fact that a person's profile or views may have come to the attention of their national authorities will lead to any adverse reaction giving rise to a real risk of serious harm.

(e) In assessing whether the foreseeable reaction of the authorities in the country of origin would be because of an actual, or attributed, Convention reason, such as political opinion or religion. [21]

(f) An individual who was a previously active blogger in their home country may cease posting in the country of refuge out of fear for their family members, or they may start posting under an anonymous name.

(g) An asylum seeker with no genuine political views and no prior social media presence begins to publish posts which are inconsistent with their previous behaviour.

10. However, just because we live in the age of social media does not necessarily mean that every asserted event will find some form of confirmation online, and courts have found an assumption to the contrary unwarranted. [22] On the other hand, the deletion of an asylum seeker's social media account in the course of status determination may raise an inference that their credibility is suspect. [23]

11. Social media evidence may arise in the context of any kind of international protection claim, though for ease of reference our examples will concentrate on claims arising from political opinion. The evidence may relate to a continuum of scenarios:

(a) A genuine political activist with a historic profile abroad may continue their blogging and posting, wholly consistently with their past conduct.

(b) An individual with a modest profile abroad historically (below the level that might have otherwise attracted persecution) might begin posting their views in the country of asylum, because that is the only way of conducting any political activity once they are no longer in their country of origin.

(c) A previously non-demonstrative individual might respond to an environment of free speech by posting views that they had not previously expressed. This would be a response to what is often called the chilling effect of repressive laws in the country of origin.

(d) An asylum seeker's account of their personal history may be undermined by the contents of their social media profile or messages to third parties (or it may be supported by such material). [24]

(e) An asylum seeker whose claims as to their personal history are rejected as wholly dishonest such that the conclusion must be that their current social media activities are not underlain by any genuine beliefs.

(f) An asylum seeker with an alleged history of public and published political activities may have stopped posting their views online once they have arrived in the country of asylum, casting doubt on whether they continue to hold the beliefs they previously asserted.

Legal framework

12. Social media may be new, but the underlying themes it raises are not. Most of them are addressed in UNHCR's long-standing Handbook on Determining Refugee Status (which whilst directed at the Refugee Convention, gives guidance which is often relevant to subsidiary protection claims too). [25] The Handbook states:

"83. An applicant claiming fear of persecution because of political opinion need not show that the authorities of his country of origin knew of his opinions before he left the country. He may have concealed his political opinion and never have suffered any discrimination or persecution. However, the mere fact of refusing to avail himself of the protection of his Government, or a refusal to return, may disclose the applicant's true state of mind and give rise to fear of persecution. In such circumstances the test of well-founded fear would be based on an assessment of the consequences that an applicant having certain political dispositions would have to face if he returned. This applies particularly to the so-called refugee "sur place".

…

95. A person becomes a refugee "sur place" due to circumstances arising in his country of origin during his absence. ..

96. A person may become a refugee "sur place" as a result of his own actions, such as associating with refugees already recognized, or expressing his political views in his country of residence. Whether such actions are sufficient to justify a well-founded fear of persecution must be determined by a careful examination of the circumstances. Regard should be had in particular to whether such actions may have come to the notice of the authorities of the person's country of origin and how they are likely to be viewed by those authorities."

13. We can see in this foundational UNHCR guidance a number of the issues that have subsequently reverberated through refugee status determination:

(a) Opinions may only become apparent after leaving the country of origin.

(b) Assumptions may be made by the authorities abroad simply based on claiming asylum.

(c) Relevant considerations are likely to be whether there has been verifiable association with dissidents or acts of political expression, and the reaction of the authorities abroad.

14. Even acts undertaken in bad faith may create risk. Regional and national courts have found that there is no exclusion clause for acting in bad faith, and so an asylum seeker may have a viable refugee claim based on expressions of political views that they do not truly hold. [26] However such cases should receive especially close examination: in its intervention in one of them UNHCR suggested that "a more stringent evaluation" [27] would be appropriate.

15. The European Union's Qualification Directive also addresses the possibility that sur place activities have been conducted with a view to manufacturing an international protection claim: [28]

"The assessment of an application for international protection is to be carried out on an individual basis and includes taking into account: …

whether the applicant's activities since leaving the country of origin were engaged in for the sole or main purpose of creating the necessary conditions for applying for international protection, so as to assess whether these activities will expose the applicant to persecution or serious harm if returned to that country;

A well-founded fear of being persecuted or a real risk of suffering serious harm may be based on activities which have been engaged in by the applicant since he left the country of origin, in particular where it is established that the activities relied upon constitute the expression and continuation of convictions or orientations held in the country of origin."

Judicial decisions

16. One national Tribunal [29] itemised the following relevant factors to be considered when assessing risk on return having regard to sur place activities, here in the context of public participation in demonstrations within sight of an asylum seeker's country of origin's Embassy or Consulate. It emphasised:

(a) the nature of the activities in question (a demonstration's political objectives, the role of the asylum seeker as a leader, mobiliser, organiser, crowd member or banner carrier, the regularity of these activities, the publicity attracted, and how matters will be seen by the regime abroad);

(b) identification risk via evidence (if it can be reasonably expected) of surveillance of demonstrators via filming or crowd infiltration by agents, and the foreign regime's capacity to identify individuals via advanced technology such as facial recognition or allocating human resources to fit names to faces in the crowd;

(c) factors triggering inquiry/action on return such as political profile and category of activities, immigration history as to mode of departing the country of origin which may lead to the necessity of the country of asylum negotiating directly with the country of origin over emergency travel documents;

(d) the consequences of identification, ie whether there is differentiation between demonstrators depending on the level of their political profile adverse to the regime; and

(e) identification risk on return: is positive identification systematically stored and used and accessible to border posts?

17. Any legal process will focus on the available evidence. But national courts have warned that expectations should not be set too high: [30] intelligence services worthy of the name are likely to mask their surveillance activities. Accordingly it may be appropriate to assume, regarding protection claims brought in relation to States known to suppress dissent and to take an interest in their reputation abroad, that public demonstrations in host countries will be filmed or photographed, and that they may well have informers amongst expatriate communities. A common-sense approach should be taken to assessing whether it is realistic to expect corroborative evidence.

18. When considering the likely meaning to be attributed to social media postings, it may be appropriate to bear in mind the nature of the medium. Judges have identified the format as a conversational medium for which an impressionistic approach is fitting, bearing in mind that very often the intention of the "App" and of the post is that the message be passed on to other users. [31] On the other hand, cynical postings online may attract close scrutiny and undermine a claim. Depending on the evidence, national security services may be assumed likely to concentrate on persons of genuine political interest rather than on opportunists.

19. International protection claims involving expression of genuine political opinions will require consideration of the legal consequences of an asylum seeker having to conceal their true beliefs on a return to their country of origin in order to avoid serious harm. As one judge put it in the related context of concealment in gender preference claims: [32]

"The way he conducts himself may vary from one situation to another, with varying degrees of risk. But he cannot and must not be expected to conceal aspects of his sexual orientation which he is unwilling to conceal, even from those whom he knows may disapprove of it. If he fears persecution as a result and that fear is well-founded, he will be entitled to asylum however unreasonable his refusal to resort to concealment may be. The question what is reasonably tolerable has no part in this inquiry."

20. The UK's Upper Tribunal has reviewed some issues arising from the use of social media evidence. [33] It had the benefit of receiving answers to written questions from Facebook Ireland and the input of two expert witnesses, a research assistant at the computer laboratory of the University of Cambridge, and the director of research of a campaigning and advocacy organisation for the rights of those wishing to publish social media material, without hindrance, in Iran and the wider Middle East. The Upper Tribunal accepted that both witnesses had relevant expertise and noted their evidence that:

(a) Permanent deletion of an individual user's Facebook account was possible and would remove everything added to Facebook, from that user's perspective.

(b) However that deletion might well take up to 90 days to take full effect albeit that it would not be accessible to individual users over that period.

(c) Deletion was partial rather than comprehensive: for example messages between users would not be deleted, because of their storage separately in user inboxes; photographs shared with other users would remain extant within the latter's accounts.

(d) Facebook would disclose information from users' accounts to governments in certain circumstances (Facebook reports on this via regular Transparency Reports): [34] it would:

"comply with government requests for user information only where we have a good-faith belief that the law requires us to do so. In addition, we assess whether a request is consistent with internationally recognized standards on human rights, including due process, privacy, free expression and the rule of law. We scrutinize every government request we receive to make sure it is legally valid, no matter which government makes the request. When we do comply, we produce only information that is narrowly tailored to respond to that request. If we determine that a government request is deficient, we push back and engage governments to address any apparent deficiencies. Where appropriate, we will legally challenge deficient requests. A Mutual Legal Assistance Treaty request or letter rogatory may be required to compel the disclosure of the contents of an account (see Our Continuing Commitment to Transparency)."

(e) If Facebook considers content is illegal under local law, then it is made unavailable in the relevant country or authority.

(f) In general, internet search engines may search online and cache information (there is a privacy setting that can prevent this, but most users opt for the defaults: ie fully public, or shared with friends only).

(g) The modern approach of intelligence services to social media would probably not be to conduct enquiries at the border when an asylum seeker disembarks. Rather they are likely to "scoop" vast quantities of data periodically and search within that downloaded resource. Web crawler software can "scrape" data from many websites. Well-known and established Apps such as Facebook are alive to this possibility and aim to prevent it. It appears that web crawlers can only obtain limited information from Facebook at present. There was no evidence that Facebook had itself been "hacked".

(h) There are various ways to access a person's Facebook records – simply by engaging with an account which is public in the first place, or seeking to obtain access to non-public material via false friend requests (ie requests from user accounts asserting sympathy with the asylum seeker's cause but in truth emanating from their country of origin's security forces), [35] or tricking them into revealing their log-in details (eg by phishing and spear-phishing, by which personal information is sought on false pretences with a view to obtaining or securing the means to access a password).

21. Having regard to that evidence, the Upper Tribunal in XX Iran concluded:

(a) Intelligence services are likely to place significant weight on a person's "social graph": i.e. their position within a broader network of related people and their relative importance within that network. This will likely determine the level of surveillance they subsequently receive.

(b) It was possible for intelligence services to extract information about users even if they kept their accounts private: because their "Friends" with public settings on their own accounts might have republished the material.

(c) An asylum seeker's prominence via social media could usually be assessed by the level of interaction with their account: e.g. via "likes" and "comments".

(d) Oppressive regimes were likely to research the profile of a returnee at the moment an emergency travel document was sought on their behalf, rather than at the border on return.

(e) Social media evidence from Facebook should, for transparency, be presented via the "Download your information" function which would give a full overview of the account and avoid only partial disclosure.

(f) One aspect of status determination should be to consider whether a returnee would, if faced with imminent return to their country of origin, close their Facebook account, and not reveal its prior existence if questioned by their national authorities on return. This could reasonably be expected of individuals who lacked any genuine political beliefs and where the social media usage in question was not linked to the exercise of any fundamental right. [36]

Presentation and consideration of social media evidence

22. Sometimes it may be considered desirable to access an asylum seeker's social media accounts with a view to seeing if the material therein corroborates (or undermines) the case being put. For example, when a young person's age is disputed, their postings and their communications with friends and family might cast light on their maturity. Where the national legal regime includes a duty of candour by which the parties must place all their cards on the table it may be appropriate or necessary for the lawyers representing children to analyse and disclose relevant social media messages to the court.

23. In so doing legal representatives would need to ascertain what social media Apps are used, identify the content therein, and consider whether anything adverse to their client needs to be disclosed. A disclosure statement explaining the steps taken might be appropriate explaining the scope of the research and that the work undertaken was reasonable and proportionate. [37] For example, in the cases of young persons and children, disclosure of social media accounts is likely to represent an interference with private life and so excessively broad disclosure would not necessarily be appropriate (eg a demand to hand over one's passwords to a government legal team would be likely to be disproportionate). In those jurisdictions where the legal principle of proportionality applies, it will be appropriate to consider the necessity of making a disclosure order against an asylum seeker and whether a less intrusive measure is appropriate. [38]

24. The appropriate means of presenting social media evidence needs to be considered. National procedure rules may address the issue directly or by implication. Courts and tribunals have varying jurisdictions: some are limited to points of law. But judges with power to consider the facts may wish to check the online version at the internet domain address, either directly or indirectly depending on whether their jurisdiction is inquisitorial or adversarial in nature. But it will usually be desirable for a hardcopy (or its equivalent in the modern era of electronic filing) to be available so that the legal representatives for the government and the asylum seeker as well as the judge can ensure they are literally "on the same page" when reviewing the evidence.

25. Screenshots of posts can be falsified or presented selectively and thus only represent one component in a larger body of evidence. It may be thought desirable for the asylum seeker's legal representatives to comply with a form of standard directions as to how best to present social media evidence: for example confirming that the critical pages are public and have been so at key moments throughout the period during which they have represented the asylum seeker, including the period that the asylum application and subsequent appeals/reviews have been extant, and giving a concise summary of the associated metadata such as the account's following and the numbers of likes, shares and other forms of engagement. [39] This would avoid matters which are in truth objectively ascertainable with appropriate case management being left to conjecture. Case management directions may be appropriate to allow for social media evidence to be accessed and viewed in the course of a hearing.

Conclusions

26. The review above of salient legal issues arising from the social media evidence in international protection appeals highlights several recurring legal themes:

(a) Is the asylum seeker relying on social media evidence as part of the material supporting their international protection claim?

(b) Has the material been presented in a way which is comprehensive rather than selective?

(c) If so, how can this information be accessed safely and lawfully?

(d) If not, should steps be taken to encourage the disclosure of any such material? What are the privacy and private life implications of so doing?

(e) Will their activities have come to the attention of the authorities of the country of origin and, if so, what is their likely response?

(f) Should the asylum seeker reasonably be expected to desist from their social media activities?

(g) How can any social media evidence relied upon in an appeal be preserved?

Contributors:

• Mark Symes (Deputy Judge of UK Upper Tribunal Immigration and Asylum Chamber)

• Louise Moor (Judicial Officer and Head of Refugee Stream - Immigration and Protection Tribunal, New Zealand)

• Christine Cody (Judicial Registrar – Migration, Federal Court of Australia)

• Dr Sophie Roche (Head of Asylum Research Unit) and Dr Philipp Wittmann (Judge) – both at the Higher Administrative Court of Baden-Württemberg

• Hugo Storey (Judge of UK Upper Tribunal Immigration and Asylum Chamber, retired)

• Julian Phillips (Resident Judge of UK First-tier Tribunal Immigration and Asylum Chamber)

These Guidelines and Explanatory Note are the views of individual judges and do not represent the view of the courts and tribunals for whom they work.

[39] Though the more technologically advanced the regime in question, the less prominence posts will need to have for there to be a real risk of attracting adverse attention.