The Institute for Digital Rights submitted its Comments on "Impact of Artificial Intelligence Systems on Good Governance” to the Advisory Committee of the UN Human Rights Council on 5 May 2025.
The Advisory Committee, at the request of the Human Rights Council (Resolution 57/5), is conducting a study on the impact of AI systems on good governance. That study aims to
indicate in particular the areas in which artificial intelligence systems can contribute to promoting and protecting human rights through good governance and those in which such systems pose a challenge to good governance and human rights, highlight good practices around the globe on the ways to develop, deploy, use and govern artificial intelligence systems, following a risk-based approach, to promote and protect human rights through good governance and identify necessary safeguards.
https://www.ohchr.org/en/calls-for-input/2025/call-inputs-study-human-rights-council-advisory-committee-role-good-governance
The Institute for Digital Rights emphasized that, in order to harness the potential benefits of AI for good governance while preventing adverse human-rights impacts, affected individuals must be guaranteed AI explainability, interpretability, and the right to contest decisions. However, because these safeguards are often difficult to fully realize—especially with advanced methods like deep learning—it is essential to conduct ex ante and periodic human-rights impact assessments, particularly for AI in public services. In areas where AI-based decision-making may have serious human-rights implications, it is also necessary to impose obligations for documentation, record-keeping, explanation, and human oversight, and to establish independent oversight bodies empowered to investigate harms and provide remedies.
Addressing bias against social groups cannot be left solely to technology companies or technical experts; domestic and international principles and norms must be established for identifying and preventing AI-driven discrimination. States should review and improve the procedures, practices, and legislation related to anti-discrimination, and governance frameworks must include the participation of affected individuals or organizations that represent them.
However, the Korean Framework Act on the Promotion and Advancement of Artificial Intelligence, passed by the National Assembly on 21 January 2025, does not meet these standards. The Institute for Digital Rights therefore called on the Human Rights Council to clearly establish a human-rights-based approach to AI governance, and in particular to advocate for independent oversight and effective remedies for rights violations caused by AI.