London Consultations on the need for, and feasibility of, an International AI Bill of Human Rights.

Written by Imogen Rivers, Dhil Scholar at the Institute for Ethics in AI  

On Thursday 10 July 2025, the Accelerator Fellowship Programme of the Institute for Ethics in AI, University of Oxford, held an expert discussion led by fellow of the programme, Professor Yuval Shany, at King's College London, on the topic of an AI Bill of Human Rights.  

Professor Shany, Hersch Lauterpacht Chair in International Law and former Chair of the UN Human Rights Committee, guided a panel of international experts in exploring whether an international AI Bill of Human Rights is needed.  

He was joined by Professor Alondra Nelson, architect of the Blueprint for the AI Bill of Rights, which underpins President Joe Biden’s executive order on AI issued in October 2023, and fellow of the Accelerator Fellowship Programme, and by Dr Joy Buolamwini, fellow of the Accelerator Fellowship Programme, founder of the Algorithmic Justice League, and author of the bestseller Unmasking AI. 

Evaluating existing regulations 

Professor Shany began by critically examining the prevailing assumption that existing human rights treaties are sufficient to address emerging challenges presented by artificial intelligence. While acknowledging the significant risks AI poses to established human rights, particularly trust, privacy, and autonomy, he also highlighted the transformative potential AI holds, especially in medicine and transportation. 

The discussion moved to a critical analysis of current regulations. Professor Shany compared Article 22 of the EU General Data Protection Regulation (GDPR) 2016, which addresses automated decision-making, with Article 9(1) of the Council of Europe’s Convention 108+. He argued that these regulatory provisions, despite their intention to safeguard individuals, lack the strength, clarity, and enforcement necessary to uphold robust human rights protections. 

The panel then examined the proposed EU AI Act alongside the Council of Europe’s Framework Convention on Artificial Intelligence and Human Rights. Professor Shany suggested the EU AI Act’s specificity, and its exemptions for national security applications, undermines its efficacy as a human rights framework. Conversely, the Council of Europe’s Framework Convention was deemed too general to offer practical guidelines for protecting rights in an AI-driven world. Professor Shany concluded that, although the US AI Bill of Rights provides promising policy guidance, it alone is insufficient for addressing the global human rights challenges posed by AI. 

Applying human rights to AI 

The event further considered how existing human rights law might adapt to AI-specific challenges. Rights to equality and non-discrimination were identified as essential for addressing algorithmic biases and ensuring distributive fairness. Privacy rights, traditionally protecting communication content rather than metadata, were highlighted as inadequate in the context of AI's expansive data collection and surveillance capabilities. 

Professor Shany underscored critical gaps in existing human rights frameworks, notably regarding transparency and explainability, fundamental values crucial to accountability and individual autonomy in the digital age. He advocated for the recognition of new human rights, including a right to data integrity to protect against the misuse of ‘deepfakes’, and a right to human interaction in automated decision-making processes, safeguarding dignity and fair treatment. 

A central question raised during the discussion was whether traditional human rights frameworks, designed primarily to limit government power, are adequate in regulating powerful private actors who increasingly dominate the digital sphere. 

Toward an international AI Bill of Rights 

In her remarks, Professor Nelson stressed the urgency of reaffirming human rights principles in the face of narratives that frame AI as fundamentally incompatible with existing rights frameworks. She argued for proactive engagement, building robust international standards to counter this perspective effectively. 

The event concluded with a vibrant exchange on topics such as the inadequacies of cookie-consent agreements in safeguarding rights, the superficiality of current explainability standards, and the need for accountability throughout technology supply chains. Participants reflected critically on prevailing assumptions of technological ‘progress’. 

A closing reflection 

Dr Joy Buolamwini closed the event with her poem, ‘To the excoded’. Her work underscores the human cost of algorithmic discrimination, stressing the imperative for a rights-based framework ensuring dignity, equity, and fairness for all people in an increasingly AI-driven world. 

Following the series of global consultation events held throughout 2025, Professor Shany will publish a white paper on the topic of an International AI Bill of Human Rights as he concludes his fellowship. The white paper will be available in early autumn on the programme's website; please keep an eye on the website for further updates. 

For more information or to become involved, please contact us at aiethicscomms@philosophy.ox.ac.uk.