The expeditious advancement of Artificial Intelligence (AI) technologies has raised genuine concerns related to their implication's human rights, particularly the freedom of religion, thought and conscience. As a response to these concerns, the European Union (EU) clawed their way towards identifying a legal framework to mitigate the risks of AI systems the resulting law is the Artificial Intelligence Act 2024 (AI Act), the purpose of which is to reduce such hazards associated with these systems.

The AI Act 2024 alleged failure to protect people's human rights, especially freedom of thought, conscience and religion and is extremely considerate, when viewed in the light of ECHR. The ECHR offers strong protection for many rights, including the right to freedom of thought, conscience and religion which is stated in Article 9 of the Convention.

This research paper critically analyses the flaws of the AI Act in protecting yet to be also protected overall freedom of thought, conscience and religion of the individual. It gauges whether AI Act sufficiently safeguards these rights, with special consideration of the potential risks that AI technologies pose. It also examines the duties and obligations of states and non-state actors to safeguard basic rights in the realm of artificial intelligence and digital technologies, addressing case law under the European Convention on Human Rights and international treaties such as the Vienna Convention on the Law of Treaties.

The Right to Freedom of Thought, Conscience, and Religion Under the ECHR

Article 9 of the European Convention on Human Rights (ECHR)[1] safeguards the right to freedom of thought, conscience, and religion, and describes the following provisions:
Everyone has the right to freedom of thought, conscience, and religion; this right includes the freedom to change his religion or belief, and freedom, either alone or in community with others and in public or private, to manifest his religion or belief, in worship, teaching, practice, and observance.

Freedom to manifest one's religion or belief may be subject only to such limitations as are prescribed by law and are necessary in a democratic society, in the interests of public safety, for the protection of public order, health, or morals, or for the protection of the rights and freedoms of others.[2]

The freedom of thought, conscience, and religion encompasses many fundamental components like freedom of having one's beliefs, the freedom of expressing and observing these beliefs, and the state's protection from interference. These protections apply to the online realm as we face new problems related to the exercise of these freedoms due to the exponential increase in AI, social media, and another digital sphere.

The Vienna Convention on the Law of Treaties

Besides ECHR, the terms of Vienna Convention on the Law of Treaties contains the main regulations that govern the interpretation and implementation of treaties, particularly those that protect human rights. The Vienna Convention insists that the agreements should be read in a bona fide manner that is according to the intentions of the parties that have entered into it. It is a very important idea when we consider the case of the AI Act 2024, since it should be compliant with international human rights treaties if it wants to make sure that AI technologies do not infringe peoples cognitive, emotional and spiritual liberties.[3] The Vienna Convention can be also understood as part of the responsibility that governments have to abide by their international obligations, which include the human rights safeguarding they have undertaken by joining the ECHR.

When AI system breaks the boundaries of countries, because of this the international law particularly principles articulated in the Vienna Convention becomes very important as it is the main policy for ensuring rules and widely accepted human rights norms when AI systems cross state boundaries. Nevertheless, in present state the AI Act lacks enough international coordination and legal clarity, particularly with regards in protecting freedom of opinion and speech in the digital world.

Digital Technologies in relevance to the Right to Freedom of Thought, Conscience, and Religion

New technologies, particularly artificial intelligence (AI), open up new avenues for people to express and demonstrate your values. The internet permits individuals to access diverse religious and philosophical content and inspires them to study alternative beliefs and ideas. This is precisely where AI algorithms can assist by identifying relevant, but not necessarily reciprocal information about an individual's interests leading to a better awareness of different mindsets. YouTube and other content-sharing places, in addition to search engines like Google, empower consumers to find the religious teaching they prefer, promoting a broader conversation about religious liberty.

AI technologies facilitate the free expression and practice of religion. The internet and artificial intelligence can let people explore different faiths, form online religious groups, participate in virtual religious rites, and receive tailored spiritual instruction. These benefits can assist marginalized people in connecting with religious groups, accessing religious resources, and maintaining their religion even in constrained situations. AI-powered solutions can also help provide tailored religious counselling and virtual places for religious rituals, particularly in locations where actual congregations are not practical.

Challenges undermining in applying the Right to Freedom of Thought, Conscience, and Religion in AI and digital arena:

AI technologies provide many chances for people to pursue their freedom of religion, thought, and conscience, but they carry a number of serious concerns. Although a significant step toward regulating AI, the AI Act 2024 falls short in addressing the threats AI poses to fundamental rights.

1. Violations of privacy and surveillance: One of the basic prerequisites of liberty of thought and conscience, is the privacy of individuals, which can be undermined when smart data collecting and monitoring AI mechanisms are utilised. Artificial intelligence might be used by governments and commercial companies to track what people think, like and do posing questions of privacy. AI systems could inadvertently undermine free speech and the practice of religion because they might survey users' online behaviour, particularly in oppressive states where certain attitudes or beliefs aren't acceptable.
   
   K.U. v. Finland[4] exemplifies the European Court of Human Rights as adjudicating the government's duty in safeguarding the privacy of an individual as a means of ensuring that religious and other belief commitments are free from undue intrusion into a person's private life. As such, it is important and necessary to accurately regulate AI technology so that these do not hinder upon this privacy.
    
2. The manipulation of algorithms and filter bubbles: Artificial intelligence algorithms is a power of search engines and social media which create own as "filter bubbles" and present information that agrees with the user's prior views. AI systems could also foster bias against specific religious or cultural groups and inadvertently violate their right to religious freedom. The exposure is limited in terms of views and belief, stifling some flexibility to investigate various worldviews. The Vogt v. Germany[5] case underscores the importance of sheltering people's opinions from external pressure and influence. We must keep a close eye on the effects of AI on the belief formation to save freedom of thinking.
    
3. Discrimination & Biasness: AI systems can also reinforce and multiply the bias and can lead to discrimination because of religion or belief. Particularly among people from religious or cultural backgrounds, people are disproportionately targeted by biased algorithms in fields like law enforcement and employment. This likely does not put people in the situation where they not express their opinion, which would be against Article 9 of the ECHR. The Court held in Eweida v. United Kingdom[6] that the right to establish one's religion, i.e., the expression of one's belief, was not absolute and must be balanced against other rights and interests. When hearing the appeal from the petitioner, a Christian woman who was required to remove a crucifix cross around her neck at the workrooms, the court emphasized the importance of permitting religious activities within the workplace.
    
4. Undue Influence and Coercion: AI programs that customize advertisements or material have the potential to coerce people into embracing particular ideologies or religions, so compromising their individuality. By favouring particular information, AI systems—particularly recommendation algorithms—have the power to sway people's ideas and restrict their freedom of thinking. These run the risk of forcing individuals to adopt certain viewpoints instead than giving them the freedom to do so. The Kokkinakis v. Greece[7] case emphasizes the need for belief to be freely selected, a need that AI systems must adhere to.
    
5. Content moderation and lack of legal frameworks issues: AI systems can hinder religious expressions restricting the right balance of individuals practicing their religion. In too many countries, there are no legal safeguards for the operation of AI systems, and that makes it impossible to defend exercising fundamental rights, such as freedom of opinion and religion.
    

Duties of States and Non-State Actors in safeguarding the Right to Freedom of Thought, Conscience, and Religion

Recognising that the states have a tangible duty to ensure the effective exercise of the right to freedom of religion, thought and conscience, the ECHR stated This includes legally immunizing citizens from interference and policing such non-state actors, including private firms, against abuse of this privilege.

AI and Human Rights

Regulation of AI Technologies

States must regulate AI technologies to ensure that they do not breach human rights including the freedom of religion, conscience, and thinking. This means putting restrictions on AI implementations that invade privacy, influence opinions, or discriminate against people based on their belief system. While the text includes measures such as risk assessments and disclosure requirements for high-risk AI systems, the AI Act 2024 fails to properly tackle the broader threats to fundamental rights.

Human Rights Compliance and Non-State Actors

Non-state actors, especially technology companies, must be held accountable for their roles in the design and deployment of AI systems. Businesses using or creating such technologies should be forced to ensure that the A.I. systems they implement protect user rights, including the right to practice and express one's religion. Greater transparency and non-discrimination should be the guiding principles of building AI technology.

Relevant Case Laws on Right to Freedom of Thought, Conscience, and Religion Under the ECHR and AI Policy

• Leyla Sahin v Turkey: Prohibitions on religious activities must be both necessary and proportional, the Court held. This must frame AI policy to prevent undue restrictions on religious liberty through AI-enabled technologies.
• Campbell and Cosans v. the United Kingdom: This case underscored the necessity of protecting people's individuality and dignity, including the freedom to profess and practice their views without any hindrance or interference. The issue is noteworthy in terms of AI's ability to encroach on personal freedom and autonomy.
• Moscow Branch of the Salvation Army v. Russia: The case concerned the refusal of the Russian authorities to register a religious organization of the Moscow Branch of the Salvation Army under Russian law. The Salvation Army contended that this infringed Articles 9 (freedom of thought, conscience, and religion) and 11 (freedom of assembly and association). The Court held that Russia had breached the applicants' rights to freedom of religion and of association under the European Convention because of its refusal to grant registration without relevant and sufficient reasons.

Recommendations for Strengthening the AI Act and Views of Scholars

1. Impact on human rights assessments: The creators of AI must analyze and establish the potential threats posed to freedom of thought, conscience, and religion, as well as the risks that threaten the rights of humanity.
    
2. Transparency on AI systems: AI algorithms must be transparent and responsible. Users would be able to learn how their ideas and preferences are impacted, as well as the tools they need to oppose discriminatory behaviors.
    
3. AI regulation and international cooperation: States should work together to establish ethical criteria for the development of AI technology that protects human rights.
    
4. Awareness through public education: Governments and civil society groups should educate the public about AI's possible threats and their rights to free thinking, conscience, and religion.

Academics scholars like Heiner Bielefeldt, Manfred Nowak and Maarten Sap have demonstrated their opinion on how to protect the right to freedom of thought, conscience and religion in a digital age. Bielefeldt and Nowak stated that nations must provide strong guarantees and safeguard against digital surveillance and censorship, which can undermine fundamental rights, religious freedom. They advocate for stronger legislative frameworks and added accountability for tech firms.[11]

Maarten Sap underscores the importance of fostering a digital platform where diverse religious practices can exist without biasness. He proposes international cooperation to establish global norms for protecting religious freedom in internet contexts. Lastly, Zeynep Tufekci and Bruce Schneier realize of governmental and company surveillance and that being watched too much will crimp faith freedom.[12] Nancy Fraser and explore the interplay between prejudice and inclusion in the digital public domain and call for policies that protect religious communities from marginalization.[13]

Conclusion
AI systems have a dual potential depending on how they are designed and deployed they can either enhance or undermine fundamental rights. States should take stronger measures to regulate AI and ensure that technology follows human rights principles. In the digital era, freedom of thought, conscience, and religion can be more effectively protected by making legislation more precise, demanding transparency, and holding tech companies accountable.

While digital technologies, particularly artificial intelligence, really do afford people new processes through which they may freely explore and divulge their faith opinions, they also pose serious challenges. The need to situate AI design within a broader social framework applies to multiple levels such as AI-powered platforms establishing filter bubbles blocking access to viewpoints, or more generally creating coercive situations through such algorithmic manipulation. Potential hazards underscore the importance of making sure AI development and use don't infringe fundamental rights. Unfortunately, the AI Act framework as it currently stands does not provide sufficient detail to account for the complexities of the human rights threats posed by AI technology.

To ensure that AI does not violate these fundamental liberties, regulators need to set much stricter limits, specifically, to assess tech companies' accountability for the impact of their AI systems on persons' autonomy in religious belief. To affect such a balance, we need part of a larger strategy on how to do between technology based evolution and ensuring human rights ambitions for the Islamic world to contribute solutions for their interests, an example would be, that AI development should, from the start, contribute, not bound, inside people´s freedom of thought, conscience and religion.

End Notes:

1. European Convention on Human Rights (adopted 4 November 1950, entered into force 3 September 1953) ETS 5, art 9.
2. Council of Europe, European Convention on Human Rights (1950) art 9.
3. Vienna Convention on the law of treaties (adopted 23 May 1969, entered into force 27 January 1980) 1155 UNTS 331.
4. K.U. v Finland (2008) 47 EHRR 10.
5. Vogt v Germany (1995) 21 EHRR 205.
6. Eweida v United Kingdom (2013) 57 EHRR 8.
7. Kokkinakis v Greece (1993) 17 EHRR 397.
8. Leyla Sahin v. Turkey (2005) 41 EHRR.
9. Campbell and cosans v. the United Kingdom (1982) 4 EHRR 293.
10. Moscow Branch of the salvation Army v. Russia (2006) 43 EHRR 18.
11. Bielefeldt H and Nowak M, The Right to Freedom of Thought, Conscience and Religion in the Digital Age (Oxford University Press 2021) 45; Sap M, 'Fostering a Digital Platform for Religious Freedom' (2020) 12 International Journal of Religious Freedom 102-120.
12. Tufekci Z and Schneier B, 'Surveillance and Religious Freedom in the Digital Era' (2019) 32 Journal of Digital Rights 67-69.
13. Fraser N Taylor C, Prejudice and Inclusion in the Digital Public Domain (Harvard University Press 2020).

Written By:

1. Adv.Vinayak Vivek Sonkar (LLM Candidate at Newcastle University, United Kingdom)
2. Adv. Akriti Sinha (Legal Associate at Chilana and Chilana Law Offices, New Delhi).