What happened
During his final address to the United Nations General Assembly, Secretary-General Antonio Guterres formally called upon member states to implement international regulations for artificial intelligence. He emphasized the necessity of creating a multilateral AI risk management framework, which he proposed should be supported by credible, independent oversight mechanisms to address the systemic risks posed by the technology.
In his final UN General Assembly speech, Secretary-General Antonio Guterres urged world leaders to prioritize the regulation of artificial intelligence. He specifically called for the development of a multilateral framework designed to manage AI risks, emphasizing that such a system must be underpinned by independent and credible oversight.
The address serves as a capstone to Guterres' tenure, during which he established a 39-member advisory body on AI in 2023. This body includes a diverse group of stakeholders, ranging from government officials to technology executives from nations including the United States, Russia, and Japan.
The speech precedes a planned briefing by the UN Security Council, where industry leaders are expected to discuss the implications of powerful AI technologies. Guterres' remarks are intended to frame these discussions around the necessity of global cooperation rather than isolated national policies.
Why it matters
Guterres' appeal marks a significant push for standardized global governance in an era of fragmented national policies. By advocating for independent oversight, he highlights the limitations of self-regulation by private entities. This call sets a high-level diplomatic agenda for the UN Security Council, which is scheduled to host AI industry leaders for briefings, potentially shifting the focus from voluntary corporate commitments to binding international standards.
The call for global regulation is significant because it challenges the current trend of disparate, state-level AI policies. By pushing for a multilateral framework, Guterres is attempting to establish a baseline for safety and accountability that transcends borders, which is essential given the global nature of AI development and deployment.
The emphasis on 'independent oversight' suggests a move away from the industry-led safety models that have dominated the sector. If adopted, such a framework could fundamentally change how AI companies are held accountable for systemic risks, moving the conversation toward international legal standards.
The timing of this call, occurring during the UN's 2026 session, places immense pressure on the Security Council to move beyond rhetoric and toward actionable governance structures that can address the rapid pace of AI advancement.
Interactive Mechanism: How It Actually Works
Explore the underlying technology behind this development interactively.
Impossibility results in algorithmic fairness (e.g. Kleinberg et al., Chouldechova) show what?
What to watch next
The primary focus is the upcoming UN Security Council briefing with AI industry leaders. Observers should monitor whether this session results in concrete proposals for the 'multilateral AI risk management framework' Guterres described. Additionally, the transition to a new Secretary-General will be critical, as the future of the UN's 39-member advisory body on AI and the continuity of Guterres' governance initiatives remain uncertain.
The immediate next step is the UN Security Council briefing involving AI industry leaders. The extent to which these leaders engage with the concept of 'independent oversight' will be a key indicator of industry willingness to accept international regulation.
The transition of the Secretary-General position is a major unknown. It is currently unclear whether the incoming leadership will maintain the same level of focus on or if the momentum generated by the 39-member advisory body will be sustained.
Watch for any formal proposals or draft frameworks emerging from the UN's advisory body, as these will provide the first concrete look at what 'multilateral risk management' might entail in practice.