UN Human Rights Council adopts resolution on AI safeguards and accountability

The UN Human Rights Council has adopted a resolution on artificial intelligence and human rights, setting out measures for states and businesses to address risks throughout the AI life cycle. The resolution was adopted without a vote during the Council’s 63rd session.

The text recognises AI’s potential to support human rights and sustainable development, while identifying risks including discrimination, privacy violations, surveillance, information manipulation and the loss of human control over highly autonomous systems.

It calls on governments to take human rights into account when developing AI laws, regulations and policies. Relevant principles include equality and non-discrimination, transparency, privacy, accountability, accessibility and human oversight. States are also encouraged to conduct human rights due diligence and impact assessments proportionate to the nature and level of risk.

The resolution addresses how AI is used in decisions that affect people’s rights. Where AI plays a material role in such decisions, it calls for meaningful human oversight and clear accountability. Affected people should be able to obtain information about decisions, request human review and access effective remedies.

It also urges states to refrain from, or stop, using AI applications that cannot be operated consistently with international human rights law or that pose undue risks, particularly to people in vulnerable situations.

Businesses involved in the AI life cycle are reminded of their responsibility to respect human rights under the UN Guiding Principles on Business and Human Rights. This includes conducting due diligence and taking steps to prevent or mitigate adverse impacts.

A further focus is the gap between countries in their ability to develop and use AI. The Council calls for international cooperation and capacity-building, including technical assistance, knowledge sharing, research cooperation and workforce skills development. It also encourages access to digital public infrastructure, open AI models, datasets and tools, alongside appropriate human rights safeguards.

The resolution establishes two follow-up steps. The Council will hold a panel discussion at its 65th session on placing human rights at the centre of AI, focusing on capacity-building and bridging the AI divide. The Office of the UN High Commissioner for Human Rights will then prepare a report for the Council’s 66th session.

The resolution connects AI governance with existing international human rights obligations. Its implementation will depend on how governments translate the principles into laws, policies and oversight mechanisms, and how businesses apply human rights due diligence in practice. The planned discussion and report will provide further opportunities to examine capacity gaps and the measures needed to protect rights as AI is developed and deployed.