UN Photo/Elma Okic. The UN Global Digital Compact aims to bring governments and industry together to ensure that technologies like AI work for all of humanity.
UN High Commissioner for Human Rights Türk stressed today at the Generative AI and Human Rights Summit that human rights-based governance of AI risks must be strengthened, while promoting responsible business practices and ensuring that companies responsible for human rights harm are held accountable.
Türk began by noting that the 75th anniversary of the adoption of the Universal Declaration of Human Rights comes at a time of rapid change in the field of artificial intelligence, "which is a major historic moment".
But he admits that the emergence of generative AI is affecting everyone and creating a paradox of technological progress: on the one hand, "the way we live, work and solve some of the most complex challenges could be radically changed", while on the other hand, there is a risk that human dignity and rights could be further undermined.
In particular, Türk expressed concern about the ability of digital technologies to reshape society and influence global politics. He said that nearly 70 elections are planned for 2024, covering countries where 4 billion people live around the world, and that disinformation and propaganda on digital platforms could affect them. In the face of disinformation and manipulation, he stressed, "we must erect an impregnable pillar."
A comprehensive assessment of the affected areas
He further noted that there must be a comprehensive assessment of the multiple areas where AI could have a transformative impact, including potential threats to non-discrimination, political participation, access to public services, and the erosion of civil liberties.
In 2019, OHCHR established the Vital Business and Human Rights Technology Project (B-Tech Project), which aims to provide an authoritative roadmap for applying the UN Guiding Principles on Business and Human Rights to the development and use of digital technologies. On the day of the summit, the project released the Generative AI Human Rights Harm Taxonomy. Türk welcomed the move, saying that "it will contribute to a broader understanding of current and potential risks".
A framework rooted in international human rights norms
He stressed the need for a global and collaborative approach to the governance of generative AI and ensuring that the protection of human rights is at the heart of it. To this end, the widely used UN Guiding Principles on Business and Human Rights and the OECD Due Diligence Guidance for Responsible Business Conduct not only provide a solid barrier for countries and businesses, but also lay the foundation for responsible AI development.
However, he also believes that these two guidelines alone are not enough to address the challenges posed by AI, which requires a range of legal, regulatory and multilateral frameworks due to the potential for AI technology to be misused by States or criminal gangs. These frameworks must be rooted in international human rights norms, including those used to clarify the human rights responsibilities of businesses and investors.
Governance initiatives need to be coordinated
Türk cautioned that while the need for AI governance is now widely recognized, many of the current policy initiatives are inconsistent and often do not give due weight to human rights considerations, "which threatens to fragment the entire regulatory landscape."
In this regard, he stressed that B-Tech's generative AI project is a structural initiative that can provide a clearer understanding of the potential impact of AI on human rights, and provide countries and businesses with an understanding of what actions need to be taken, thus pointing the way to more coherent governance.
Remedies for human rights violations
Türk also said that countries need to protect everyone from human rights abuses caused by AI, and that companies need to ensure that their AI algorithms, operational processes, and business models respect human rights.
At the same time, OHCHR continues to urge the provision of effective remedies for victims of human rights violations by AI. Türk stressed that tech companies must recognize their responsibilities and the societal benefits of establishing effective redress regimes, and that at the end of the day, States have a fundamental responsibility to ensure redress for human rights harms, including the need to impose appropriate action on companies.
Türk concluded by calling for the solidarity of states, businesses, civil society and individuals with a common mission: to ensure that AI is in the best interests of humanity and to build a future world with humanity where technology works not only for the rich and powerful, but also for the universal advancement of human dignity and rights.