Predicting such a decision, the Commissioner issued a statement expressing concern about the possibility of restricting access for specific groups of Ukrainians, such as men of conscription age. The Commissioner also highlighted that the current safety realities in Ukraine do not meet the conditions for a safe and dignified return.

The Commissioner’s concern is not unfounded, remembering the established case law of the European Court of Human Rights (the Court). The Court has consistently emphasised that exposure to a real risk under Article 2 (right to life) or Article 3 (prohibition of torture) requires an individual assessment, rather than presumptions based on membership of a particular social group: see Soering v United Kingdom, Saadi v Italy, and J.K. and Others v Sweden.

The risk to people’s lives is very high across Ukraine and affects all citizens, but for men aged 25 to 60 it is highest because of military conscription. Moreover, as I argued in a previous Blog post, even internationally recognised and constitutionally enshrined guarantees and standards of conscientious objection in Ukraine are ignored not only by the government but also by the Supreme Court. Even if we assume that some Ukrainian men voluntarily serve in the military to protect the country, compulsory military service, in the author’s opinion, is opposable in principle as it violates the right to life.

According to the official website of the Ukrainian Ombudsman, one of the governmental human rights protectors appointed by the Ukrainian Parliament, human rights violations by military officials responsible for mobilisation in Ukraine have become systemic and widespread. The Ukrainian Ombudsman also noted that illegal detentions and beatings have become a trend that cannot be stopped without a tough response from the state. The Ombudsman further observed that these incidents are not merely individual violations but frank manifestations of violence and abuse of power by military officials in Ukraine.

Now, European bureaucrats are facing a difficult dilemma. On the one hand, under pressure from the Ukrainian government, which needs soldiers for the war, the EU is going to limit Ukrainian conscription-age men from receiving protection in Europe, which may cause many court cases about Convention violations in the near future. On the other hand, if the EU continues allow all Ukrainian men to enter Europe and receive full protection, it may increase social disagreement in Europe and undermine consensus on supporting Ukraine. This might increase the real risk of Ukraine’s defeat because of a lack of consolidated European support, which may leave Europe as the next target of Russian aggression after Ukraine.

That is why European leaders will face a very tough dilemma, but should think twice before implementing the mentioned decision.

In the author’s opinion, in this case, European governments should follow international and European human rights standards, such as the internationally recognised right to conscientious objection, the right to life, or the ban on torture, which are violated by Ukrainian authorities and should be guaranteed in the EU. Otherwise, this may be a negative signal to European men that their rights may also be violated if EU authorities decide so, justifying it by military needs rather than human rights standards.

Facts

C.M., a teacher with a disability, requested a transfer to the province of Catanzaro as part of the Italian public school staff mobility programme for the 2018/2019 school year. The scheme prioritised transfers for individuals with certain (degrees of) disabilities among the available positions. Requests for transfers within the same province were (mostly, § 20-21 & 28) granted before those for transfers between provinces.

CM’s request was denied as all vacant positions in Catanzaro had been filled through internal transfers. The Italian court concerned asked the Court two preliminary questions: (1) whether the priority to internal transfers violated the obligation to provide reasonable accommodation (Article 5 Directive 2000/78); and (2) whether this arrangement constituted (unjustified) indirect discrimination based on disability (Article 2 Directive 2000/78)?

Reasonable Accommodation

First, the Court considered the definition of reasonable accommodation, interpreted in light of Article 2 of the UN Convention on the Rights of Persons with Disabilities (CRPD) (§ 50-51). Reasonable accommodation refers to modifications and adjustments ‘where needed in a particular case to ensure to persons with disabilities the enjoyment or exercise on an equal basis with others of all human rights and fundamental freedoms’ (§ 51). In the workplace, this may include reassignment to another position (as established in HR Rail and Ca Na Negreta; § 52, 58).

The Court noted that ‘reasonable accommodation’ should be ‘understood broadly, as covering all measures that enable the needs of a person with a disability to be met in a particular case’ (§ 61). However, the Court reasoned that as the scheme provides priority automatically for those with certain (levels of) disability, it cannot constitute a reasonable accommodation (§ 61-64). A reasonable accommodation has to be based on ‘a concrete understanding of the needs specific to that worker’s situation and the difficulties and obstacles which that worker encounters in accessing and performing his or her work’ (§ 56).

Indirect Discrimination (and Positive Action)

Next, the Court considered whether the scheme could be indirectly discriminatory, i.e., liable to result in a ‘particular disadvantage to the detriment of workers with disabilities’ (§ 68-70). The scheme instead appears beneficial only to persons with disabilities (§ 73-74). However, as it distinguishes between different disabilities and the Commission had argued it to be positive action (Article 7 Directive 2000/78), the Court was also invited to consider the case from this angle (§ 78-84). It succinctly rejected this invitation, holding that it lacked information concerning the scheme’s objective (§ 82-83).

Commentary

Although the Court’s conclusion on this specific measure is understandable, it leaves broader questions unresolved. The UN Committee similarly emphasises that reasonable accommodations should be individualised and context-sensitive (General Comment No. 6, §25-26). However, focusing on a measure’s automated character may overlook measures that do meet the substantive criteria for reasonable accommodations. For example, students with specific disabilities often automatically receive exam accommodations, such as extended exam time when they qualify for accommodations (see Lammens, Spinoy, and Willems). While these should be accompanied by a further individualized assessment (and should not replace this assessment), they still constitute measures to avoid obstacles indicated by a person with a specific disability and therefore ‘enable the needs of a person with a disability to be met’ in a particular context.

Automated measures could also constitute positive action. While reasonable accommodations address individuals’ needs and are mandatory under EU law, positive action targets the underrepresentation of protected groups and remains voluntary (see Waddington and Bell). Following significant rulings regarding positive action for women in employment (e.g., Kalanke, Marschall, Badeck, Abrahamsson, and Lommers), subsequent silence has created considerable uncertainty, particularly regarding other protected groups (which the Court has only briefly addressed: see Milkova on disability; Cresco Investigation on religion and belief).

Zirvatta demonstrates that the Court continues to regard positive action as a justification for discrimination (§ 81-83). Yet positive measures may be necessary, not simply justifiable, to tackle deep-rooted and systemic discrimination (see Fredman). Unfortunately, the Court missed this opportunity to provide clearer guidance.

In 2025, three major legal developments demonstrate regulators attempting to close this gap, and all three possess the same structural blind spot.

In March 2025, the UK Home Office provided new statutory guidance on ensuring transparency in supply chains under section 54 of the Modern Slavery Act 2015 (MSA). The guidance strengthens expectations for corporate disclosures, proposes a tiered reporting system, and calls for corporations to monitor their supply chains beyond Tier 1 suppliers. The guidance, however, changes the underlying legal obligation; section 54 requires the companies to publish a transparency statement. There is no independent verification mechanism, and the compliance remains self-reported. Meanwhile, in the European Union, the Corporate Sustainability Due Diligence Directive (CSDDD), significantly amended by Directive (EU) 2026/470, requires companies to identify human rights risks throughout their value chains, including to indirect suppliers where “objective, verifiable information” indicates a specific risk. The EU Forced Labour Regulation, which will be implemented in December 2027, takes an additional step, allowing customs authorities to prohibit products manufactured with forced labour from the EU market. Together, these measures comprise the most ambitious regulatory framework for supply chain accountability ever developed. Additionally, they suffer from a shared failure; collectively, they generate duties without creating the investigative instruments necessary to discharge them.

This is where open-source investigation (OSINT) comes in, not as a technological breakthrough, but as a legal necessity.

Open-source investigation leverages publicly available information, shipping and customs records, satellite imagery, corporate registry filings, procurement databases, and social media to map supply chain structures and identify indicators of exploitation invisible to conventional audit methodologies. It is not hypothetical in the case of modern slavery. In 2021, Sheffield Hallam University’s report on Xinjiang cotton demonstrated how customs and shipping data may be used to trace forced labour inputs through various levels of supply chains into the goods of big global companies. This is exactly the sort of “objective, verifiable information” the CSDDD obliges corporations to obtain, but the Directive lacks guidance on how to obtain it.

The legal argument is straightforward: Open-Source Investigation should be recognised as a legitimate investigative methodology in modern slavery enforcement regimes. It could have three approaches. First, for regulators: Open-Source Investigation may be a standard tool for customs and national competent authorities to identify at-risk products and supply chains according to the EU Forced Labour Regulation’s enforcement mechanism. Second, for companies: The amended UK MSA guidance recommends firms to map their supply chains, and Open-Source Investigation provides a scalable, economical way to do this beyond Tier 1. Third, for litigation: Cases such as Limbu v Dyson require claimants to establish what occurred inside distant, informally contracted facilities that brands routinely deny controlling. This is the evidentiary burden that conventional corporate disclosure hardly ever meets. Open-source investigation can address it directly: shipping manifests and customs records can establish a direct and documented commercial relationship between a specific factory and a brand, establishing the brand’s knowledge of that facility beyond credible dispute; satellite imagery can document undisclosed production sites operating outside corporate audit procedures; and corporate registry filings can trace the ownership structures through which liability is deliberately obscured across supply chain tiers. Each of these technologies addresses an issue that the courts in transnational supply chain conflicts are already asking, and that self-reported company compliance data cannot answer.

The UK Joint Committee on Human Rights announced in January 2025 that it will be conducting an inquiry into forced labour in UK supply chains. It published its findings in July 2025. The Committee observed that products manufactured employing forced labour are entering the UK market, that the UK risks becoming a dumping ground for products banned overseas, and that comprehensive supply chain due diligence is urgently needed. However, the report’s proposals still do not answer the central question of a standardised approach. How should businesses, regulators and enforcement agencies actually see through the supply chains they are now legally obligated to scrutinise?

Open-source inquiry is not a supplementary tool waiting to be discovered. It is the investigative system that provides practical enforceability to these duties. 

Irreversibility matters because it eliminates the argument that third states supporting Israel can separate legitimate security facilitation from unlawful practices. Once you sustain an administration that kills through racially segregated courts, you cannot claim you only support the lawful parts.

This is where the state responsibility doctrine becomes critical. Because the ICJ characterised the right to self-determination as a peremptory norm (paragraph 233 of the AO), Israel’s continued presence constitutes a serious breach engaging the customary duties codified in Article 41(2) of the Articles on State Responsibility for Internationally Wrongful Acts (ARSIWA): no state may recognise the resulting situation as lawful, nor render aid or assistance in maintaining it. The European Union’s (EU) Foreign Affairs Council’s (Council) 21 April 2026 decision not to suspend the EU-Israel Association Agreement is where these questions become concrete. This decision followed the EU’s External Action Service finding Israel in breach of the Agreement’s human rights clause (Article 2) and the enactment of the March statute.

How State Killing Makes the Apparatus Irreversible

To understand why irreversibility changes the legal calculus, it is necessary to examine what the ICJ found and how the statutes operationalise it. At paragraph 229 of its AO, the Court identified a structure: Israel’s laws “impose and serve to maintain a near-complete separation” between settlers and Palestinians, violating the prohibition on apartheid in Article 3 of the International Convention on the Elimination of All Forms of Racial Discrimination (ICERD).

The Death Penalty for Terrorists Law makes that bifurcation lethal through differential jurisdiction. Article 3 establishes mandatory death by military court for Palestinians who cause death in acts of terrorism, excluding Israeli citizens or residents. Israeli citizens face civilian courts where Article 6 imposes death only for killings to negate the existence of the State of Israel. Human Rights Watch and Amnesty International documented that Palestinian residents face mandatory death in military courts, whilst Israeli citizens face death only for a narrowly defined intent that functionally excludes nationalist killings of Palestinians.

The military courts administering death operate with a 99.7% conviction rate, based largely on torture-extracted confessions. These conviction rates matter because they demonstrate the courts are not impartial tribunals but components of the apparatus the ICJ found unlawful. When the United Nations’ Committee on the Elimination of Racial Discrimination (CERD) stated that the law perpetuates racial discrimination, it confirmed differential vulnerability to state killing along ethnic lines. Administrative detention was theoretically reversible. State-administered execution is not.

When Third State Support Breaches the Duty of Non-Assistance

The duty of non-assistance requires no proof of complicit intent. At paragraph 278 of its AO, the Court held that the duty of distinguishing dealings with Israel between its own territory and the OPT encompasses, among other obligations, the duty “to abstain from entering into economic or trade dealings with Israel concerning the Occupied Palestinian Territory or parts thereof which may entrench its unlawful presence in the territory,” and “to take steps to prevent trade or investment relations that assist in the maintenance of the illegal situation.” The Court’s phrase “may entrench” establishes a probabilistic standard. States need not prove their conduct caused a specific harm, only that it sustains Israel’s administrative capacity.

The Court found Israel’s entire presence unlawful. There is no lawful administration to sustain. Arms, intelligence, and preferential trade that sustain capacity therefore sustain the wrongful situation, within which a death regime operates.

Unlike complicity under ARSIWA’s Article 16, which requires that assistance be given with knowledge and with a view to facilitating the wrongful act, the Article 41(2) duty covers conduct after the fact that maintains the unlawful situation, with the International Law Commission considering it “hardly conceivable” that a state would lack notice of a serious breach. Any residual doubt about notice dissolved when the CERD Committee called on States Parties to ensure that their “resources are not used to enforce or support discriminatory policies and practices against Palestinians living in the Occupied Palestinian Territory.” When the EU’s External Action Service found Israel in breach of Article 2 of the Agreement, and the Council nonetheless decided not to suspend, the Union placed itself squarely within the conduct paragraph 278 prohibits. Twenty UN experts stated that suspension is “a legal imperative,” not a matter of political discretion.

What This Reveals About Metropolitan Responsibility

The Council’s decision reveals how the ICJ’s “may entrench” standard operates as burden inversion. The question is whether conduct sustains the apparatus through which apartheid is maintained and death administered, not whether governments intend to facilitate it. Once the apparatus kills, separability collapses. The courts executing Palestinians are those that the ICJ found unlawful. International legal infrastructure has constituted settler-colonial projects in Palestine, not merely recognised them.

Article 41(2) asks when metropolitan support becomes co-constitutive of the wrongful situation. The answer: when it sustains administrative capacity that kills through racially segregated courts. This tests whether the duty of non-assistance retains doctrinal content or serves merely as a rhetorical cover for sustaining colonial structures.

The case involved an employee, JB, who worked for a Catholic pregnancy counselling service in Germany. Her role involved advising women in line with Church teachings with the objective of protecting the life of the unborn child. JB was part of a team of six, four of whom were members of the Catholic church, and two who were not. JB lost her job after ending her membership of the Catholic Church, despite the fact that her personal beliefs still aligned with Catholicism, and she was prepared to continue providing counselling in accordance with Catholic teaching. The reason for leaving the church was, instead, due to a tax dispute with the local diocese, involving the imposition of a levy on persons, like JB, in interfaith marriages with high-earning spouses. JB claimed that her dismissal amounted to direct discrimination on grounds of religion or belief. The employer argued that a requirement to remain a member of the Catholic church was justified as a genuine occupational requirement of the role, which required loyalty to the Catholic church, a loyalty which was undermined by her decision to end her membership of the Church.

The case was referred to the CJEU to determine whether Employment Equality Framework Directive 2000/78 prohibits dismissal from employment in these circumstances. The CJEU ruled that, in cases where the religious ethos employer does not require all staff to belong to the Church, and where the employee has not acted in a manner that is antagonistic to the Church, the dismissal was not lawful.

In reaching this welcome, though perhaps unsurprising, decision, the CJEU reaffirmed its position in the cases of Egenberger and IR and undertook a balancing exercise which recognized and respected the legal obligations in EU law to uphold religious autonomy, both that of the counselling service to employ those who respect their religious ethos, and that of JB, exercised here through her freedom to leave the Church. At the same time, the CJEU was firm that this has to be held in balance with the principle of equality and the need to protect the non-discrimination rights of employees.

In this case, the decision was relatively straightforward. Given that others were employed without being members of the Catholic Church, the requirement imposed on the employee to remain a member of that church could not be a genuine occupational requirement. There was no sufficient link between the occupational activities and the requirement to be a member of the Church. Whilst there was a sufficient link with the requirement to be loyal to organisation’s ethos of support for the promotion of the life of the unborn child, the employee remained loyal to that ethos: there was no evidence that the employee would not fulfil the obligations of loyalty to the employer with regard to her support for its approach to pregnancy counselling. Moreover, the reason for leaving the Church related to a separate financial matter. It did not involve a public distancing from, and thus any disloyalty towards, the ethos and teaching of the Church.

The case, while a helpful confirmation of the need to balance loyalty to the employer against freedom from discrimination, thus leaves open the question of whether, had the employer been more consistent in requiring all staff to comply with a requirement of church membership, the case might have been determined differently. The Court did not challenge the autonomy of the Church to set its ethos, rather its challenge was based on the inconsistency of the Church in its application of the ethos to some and not others.

In Katholische Schwangerschaftsberatung v JB the CJEU thus reiterated its commitment to uphold the right of religious employers to impose genuine, legitimate and justified occupational requirements on staff in order to uphold religious freedom and autonomy; but it also reiterated that the principle of non-discrimination must be interpreted strictly. Although in this case, the balance between these two interests clearly fell in favour of the employee, the case is nonetheless important. It confirms the difference in the CJEU’s approach from that in the US, where religious autonomy is paramount.  The European approach, along with the approach of the IACtHR, instead achieves balance through a commitment to fact sensitive proportionality review. This does not unacceptably erode the religious autonomy of religious organisations: instead, oversight of the exercise of religious autonomy is necessary for the adequate protection of equality rights.

The first part of this blog set out the background to and outcome of Denmark’s action for annulment of the EU Directive on adequate minimum wages. Here, we delve into the Court’s reasoning for upholding the validity of Articles 4 and 6, and the rest of Article 5, before discussing the broader implications of the judgment for Social Europe.

Keeping the bulk of the Directive alive: The Court’s reasoning

The Court confirmed that Articles 4, 5 (except the annulled sections) and 6 all fall outside the exclusion of ‘pay’ from EU competence. A common theme in the reasoning related to all three provisions is the emphasis placed on state and collective autonomy for the finding that no direct interference took place.

In respect of Articles 4 and 5, the Court stressed that Member States and social partners retain broad discretion in determining a statutory minimum wage (SMW) and negotiating and concluding collective agreements. It further highlighted that the Directive contains ‘procedural’ requirements rather than prescribing specific results that determine SMW or the outcome or content of collective bargaining. In other words, what made the difference is that EU rules would not directly influence the levels of minimum wages.

In respect of Article 6, the Court found EU countries are still free to introduce (or not introduce) variations or deductions. Article 6 ‘merely requires’ any variations or deductions to comply with non-discrimination and proportionality and thus does not directly interfere in pay levels (para 106).

On the ‘right to association’, the Court took a narrow view of the exclusion as not covering measures governing the right to collective bargaining. This finding was informed, among others, by the existence of separate provisions on the rights to association and collective bargaining in some human rights instruments (such as the EU Charter and European Social Charter). As a result, it confirmed EU competence to legislate on collective bargaining under Article 153(f) TFEU, ‘representation and collective defence of the interests of workers and employers’ (para 136). However, this legal basis requires unanimity among Member States in the Council of the EU to adopt legislation, which is difficult to achieve in practice.

A finding of potentially greater significance is that ‘working conditions’ (Article 153(2)(b) TFEU) can serve as a legal basis for collective bargaining rules. The latter need only to be a means for promoting working conditions, and their effect on representation and collective defence to be ‘ancillary’ to this aim (para 136). In this case, the duty to promote collective bargaining was construed as a means for ensuring the adequacy of minimum wages (which are considered part of working conditions). Using ‘working conditions’ as a legal basis offers a crucial advantage. Aside from the approval of European Parliament, only a qualified majority is needed among Member States in the Council to adopt legislation (i.e., the support of at least 15 out of 27 states,  representing at least 65% of the EU population).

Implications of the judgment for Social Europe

The judgment can be considered a landmark moment in the evolution of Social Europe and its complex relationship with the CJEU. Its significance lies not only in what the Court decided, but equally in the path not taken: by rejecting the full annulment of the Directive, the Court avoided a major setback for (Social) Europe. It is too early to understand the longer-run implications of the judgment, but we can identify some immediate effects.

First, the judgment finally confirms the validity of most of the Directive with its core substance, innovations, and spirit intact, allowing the focus to shift to effective implementation. Second, it adopts an interpretation of the exceptions of ‘pay’ and ‘right to association’ that allows substantial scope for interventions not only in wage-setting but also in collective bargaining. In this sense, the Directive successfully tested the boundaries of what EU law is permitted to do in these areas.

Third, and very importantly, it confirmed that if collective bargaining provisions are related to the objective of promoting ‘working conditions’, a qualified majority (rather than unanimity) suffices. The same reasoning could apply to other legal bases listed in Article 153(1), such as workers’ health and safety or equality between men and women, if collective bargaining is shown to be an effective means to promoting them. This is significant. The judgment opens up legal space here that policy-makers can creatively use to strengthen the social dimension of the European project.

Conclusion

By any measure, the judgment is overall a positive step for Social Europe. Of course, the very fact that the Directive encountered such a strong challenge shows that genuine progress towards Social Europe is often an uphill battle. And while the judgment is a cause for celebration, it is important to avoid complacency. The judgment did not bring by itself any change on the ground and the implementation of the Directive is still a work in progress. But the CJEU has opened a door for progressive social experimentation and innovation in advancing Social Europe. Walking through that door, however, ultimately depends on political will matched by concrete action.

This two-part blog explores the judgment and its implications. We first outline the main provisions of the Directive, the challenge to its validity before the CJEU, and the outcome in the case. In the second part, we take a closer look at the reasoning of the Court and its implications for future social legislation.

The Directive on adequate minimum wages

Against the background of the euro-crisis, and the deregulation and austerity measures that followed it, the Directive on adequate minimum wages was one of the instruments adopted by the EU with the potential to reassure citizens that the Union can deliver better rights and social standards. It aims to improve living and working conditions, in particular the adequacy of minimum wages, and thus contribute to upward social convergence and reducing wage inequality (Article 1). It rests on three main pillars, expressed in Articles 4-6.

Article 4 requires several measures to promote collective bargaining coverage and facilitate the exercise of the right to collective bargaining on wage-setting. This includes measures to protect trade unions and employers’ organisations participating or wishing to participate in collective bargaining from interference by each other in their establishment, functioning or administration (Article 4(1)(b)), and an obligation for Member States where collective bargaining coverage is less than 80% to provide for ‘a framework of enabling conditions for collective bargaining’ (Article 4(2)).

Article 5 requires Member States with a statutory minimum wage (SMW) to establish a procedure for setting and updating it, guided by criteria that seek to ensure the ‘adequacy’ of SMW levels. While optional, the Directive mentions a ‘double decency threshold’ for minimum wages (60% of gross median wage and 50% of gross average wage) as an indicative reference.

Article 6 provides that variations or deductions in SMW rates for specific workers should respect the principles of non-discrimination and proportionality. Such workers typically include young people, workers with disabilities, and seasonal or domestic workers. The Directive thus promotes several key social rights.

The action for annulment

Even before the end of the transposition period (i.e., the period of incorporating EU directives into the national laws of EU Member States), some Member States had introduced measures responding to the Directive that led to increases in minimum wages.

In 2023, however, Denmark brought an action to annul the Directive, in whole or only Articles 4(1)(d) and/or Article 4(2). Denmark argued that it does not fall within the Union’s constitutional competence to legislate under Article 153 of the Treaty on the Functioning of the EU (TFEU). While Article 153(1)(b) TFEU provides that the EU can adopt directives on ‘working conditions,’ Article 153(5) excludes EU measures on ‘pay, the right of association, the right to strike or the right to impose lock-outs.’ The Directive, Denmark argued, concerned ‘pay’ and ‘right of association’. In his controversial advisory opinion of January 2025, Advocate General Emiliou recommended that the Directive be fully annulled as directly interfering with the exclusion of ‘pay’.

The Court’s verdict: A minimal partial annulment

By contrast, the CJEU held that only part of Article 5 could be considered a ‘direct interference of EU law in the determination of pay’ within the EU, establishing this as the criterion determining the exclusion of competence relating to pay (paras 68-70).

The Court annulled Article 5(2), which contains compulsory elements to be included in national criteria to ensure the ‘adequacy’ of SMW. This fell within the exclusion since it concerned EU requirements relating to constituent elements of SMW that would have a direct effect on their level. For the same reason, the Court annulled the last clause of Article 5(3), which specifies that, where used, an automatic mechanism for indexation adjustments of SMW cannot lead to a decrease in SMW.

While these provisions would have made procedures for setting adequate SMW more robust, their annulment does not compromise the core obligation in Article 5. Their annulment may also be of little practical significance, as Member States have already transposed the compulsory criteria in national regulations, and few Member States use an automatic mechanism for indexation.

The Court upheld the rest of the Directive. In the second part of this blog, we analyse more closely the Court’s reasons for reaching this outcome and the broader implications of the judgment.

What is interesting to note is the fact that these techniques have been subjected to a complete prohibition, and have not been made contingent on the consent of the data subject. The only exception, under Recital 29, applies to the use of medical techniques deployed with explicit consent, e.g. the use of Deep Brain Stimulation to cure neurological disorders. Such a blanket prohibition differs, for instance, from the framework for processing of special categories of personal data under Article 9 of the General Data Protection Regulation (GDPR), which can still be processed with informed consent. Assuming that the use of algorithmic systems envisaged under Article 5(1)(a) of the AI Act involves processing of data on cognitive and behavioural states of data subjects, it is apparent that the AI Act possibly carves out a special category of data derived from a person’s internal decision-making processes that is afforded even higher protection than that ordinarily available under Article 9 of the GDPR. 

The reasons for providing greater protection to data relating to a person’s cognitive and affective states appear to be obvious: as the Information Commissioner’s Office (ICO) in the UK notes (p. 19), consent in the realm of “neurotechnology” may be an inadequate safeguard when it comes to ensuring control over one’s personal data. In a normal situation, by providing consent to processing of personal data, the data subject retains conscious control over what items of personal data are being divulged. When neurotechnology itself manipulates and modifies the data subject’s cognitive and behavioural functioning by directly modulating electrical activity or artificially stimulating certain areas of the brain,  the “voluntary” and “informed” nature of consent becomes suspect. Further, reports have emerged of neurotechnological advancements that purport to have the ability to convert brain scans into transcribed speech in real time. In the face of such technology, even if initially consent is provided, one has no control over what kinds of data may be unwittingly disclosed to the outside world, let alone determine the inferences that may be revealed from them. 

From a rights perspective, the prohibition in Article 5(1)(a) is itself an expression of the unqualified protection that is afforded to unexpressed “thought” or the “forum internum” in the European Convention on Human Rights (ECHR). Under Article 9(1) of the ECHR, protection is granted to “freedom of thought, conscience and religion”. However, while Article 9(2) envisages limitations on manifestation of one’s religion or beliefs, the provision is silent on any possible limitations on the freedom of “thought”. One possible interpretation of this deliberate wording points to the fact that unexpressed “thought” or the forum internum under the ECHR is given an inviolate and absolute status, free from any potential limitations or restrictions. This can explain why under Article 5(1)(a) of the AI Act, technological advancements that tinker with “thought” at its most basic level, namely, the cognitive process of thinking, are not permitted even with consent, subject to the medical use exception. 

While technological advancements ought to be welcomed, it is imperative to ensure that the freedom to preserve one’s innermost thoughts and the autonomy to make choices are not tinkered with. Viewed from this angle, Article 5(1)(a) of the AI Act acts as a guardrail that effectively safeguards the inviolability of “thought” from neurotechnological intervention. 

The rule of law as defined by A.V Dicey refers to supremacy of the law as opposed to arbitrary exercise of powers, equality of all men before the law, and protection of individual human rights of people (Cf Lord Bingham, Joseph Raz and Brian Tamanaha). Legal instruments have long been adopted to  provide frameworks for protection of the rule of law such as the 1215 Magna Carta which introduced the right to a fair trial, protection from arbitrary detention, and the requirement that laws must be applied equally, the writ of habeas corpus securing the presence in court of the defendant or criminal suspect, and the Constitution of the United States of America in operation since 1789, a historical turning point in the development of the rule of law where it sought to create a strong central government whilst maintaining an individual’s autonomy. This later led to the American Declaration of Human Rights and impacted the adoption of the 1789 French Declaration of Man and the Citizen. Globally, there is the Universal Declaration of Human Rights, adopted by the United Nations General Assembly in 1948, which enshrines the rule of law in the international law sphere.  However, with the increasing use of AI in decision making, questions must be asked about how it affects the operation of concepts associated with the rule of law, such as transparency, fairness, and explainability, as this technology is no longer a figment of imagination.

Whilst AI plays a key role in development, its operationalisation should be hinged within the constraints of the rule of law and protection of human rights and values. Debates on AI regulation often do not take into account that AI itself has regulatory effects on the legal system which could potentially endanger the adherence to the rule of law in democratic states. The infrequency of discussions between the intersection of technology and rule of law confirm that whilst AI primarily embraces globalisation, the rule of law, for the most part, is still primarily confined to national borders. Thus, there is need for international collaboration on setting legal and regulatory frameworks to improve transparency and fairness of decisions made by AI.  As of 2023, an AI index analysis of legislative records of 127 countries shows that the number of bills containing ‘artificial intelligence’ that were passed into law grew from just 1 in 2016 to 37 in 2022. An analysis of the parliamentary records on AI in 81 countries likewise shows that mentions of AI in global legislative proceedings have increased nearly 6.5 times since 2016.

Recent developments and global initiatives that have been seen as most influential and well-known include Council of Europe’s Framework on AI, which is the first international legal instrument on AI focusing on alignment with the rule of law, and UNESCO’s G7 tool kit on AI and rule of law. Regionally, the EU Artificial Intelligence Act 2024 focuses on risk levels of AI to ensure transparency and accountability. In the US, a proposed 10-year moratorium on state level AI regulations introduced by the House Energy and Commerce Committee in May 2025, has been met with opposition from bipartisan groups and states that have regulated high-risk uses of the technology, arguing that the measure will pre-empt AI laws and regulations passed recently in dozens of states. The (Data Use and Access) Bill which is at the reporting stage in the United Kingdom’s House of Commons has raised concerns on the narrowed scope of Article 22 under the UK General Data Protection Regulation which could allow more significant AI decisions without meaningful human involvement.

By emphasising shared global responsibility, legal and regulatory frameworks can help mitigate the risks of automation bias and preserve the rule of law.  AI must operate within a clear legal framework and remain subject to judicial oversight to ensure that the rule of law is upheld.

The  EU’s AI Act 2024, which will enter into force in August 2026, has classified the use of AI profiling for border control as high risk and, as such, worthy of high regulation. Indeed, AI has been used to screen asylum and visa applications, ensure an individual is who they claim to be through biometric identification systems, and estimate illegal migration risks. Amongst these regulations is the responsibility of providers to conduct data governance, record-keeping, and the implementation of human oversight. Criminal risk profiling based solely on personality traits inferred from an individual’s data and social scoring, on the other hand, lies amongst the prohibited usages of AI listed in the Act (Chapter II, Article 5).

In addition to this, a lesser talked about risk is the use of already controversial DNA analysis for profiling. Genetic testing providers such as Ancestry.com and CircleDNA both claim to be able to assess one’s innate propensity to risk-taking, for example, based on one’s DNA (although both tests can give different results for the same person). Risk-taking is thought of as one of the main personality traits that AI could possibly be used to assess when profiling applicants  to determine their chance of irregular migration and other criminal offenses. With 23andme.com recently filing for bankruptcy after an important data breach, debates about the intersection of DNA testing, data privacy, migration, and AI are more urgent than ever, as, if not prohibited, DNA data analysed by AI could be used in some regions to facilitate migration profiling based on personality profiling. For now, however, the new practice of DNA profiling at the border in the EU, ECOWAS, and other regions, focuses on citizen identification and aims to address the missing persons and human trafficking crisis.

In contrast to this push to put some checks and balances in European border control tech, the Trump administration saw a push towards stricter immigration law enforcement, while inversely advocating for a more liberal AI use in governmental affairs and beyond. The country had indeed already passed various legislations, which vary by states, to facilitate AI profiling to prevent criminal activities whilst accounting for data privacy. In the context of the Trump administration’s migration crackdown, which has been criticised for its lack of due process, AI has been used to profile migrants based on their political inclinations, which arguably contravene several individual liberties listed in the ICCPR and beyond.

All of this could also lead to important demographic changes in the future as even just the looming prospect of a more dystopian version of the US might attract different kinds of immigrants, ideologically speaking and demographically speaking, and retain different kinds of citizens, again based on ideology and demographics. The sociological consequences of such policies on how feasible and attractive various migration destinations are could lead ethnic and sexual minorities, as well as refugees and economic migrants, to gravitate towards destinations with stronger anti-profiling interdictions, such as the EU, which could risk exacerbating the far-right migration pushback in the EU context and their Great Replacement arguments. More likely, developing and emerging destinations in the Global South (specifically the BRICS, South-East Asia, and the GCC) could also gain traction as the transition towards a multipolar global order cements.