NEW YORK, NEW YORK / RankWire.AI / – Khalilur Rahman, who presides over the 81st United Nations General Assembly, emphasized the need for a more accelerated worldwide effort to establish safeguards for artificial intelligence. During the opening of the General Debate, he expressed concern that the UN risks falling behind in the face of swift technological developments. Rahman called on governments to develop dependable protections grounded in scientific research and highlighted that AI policies should allow space for innovation. His comments underscored artificial intelligence as one of the key issues confronting the global community.
Rahman stressed that governments must take the lead in international AI regulation, but he also pointed out that experts from research, business, innovation sectors, and civil society play a vital role. He advocated for increased involvement of organizations and individuals involved in the development and deployment of AI, emphasizing that their technical expertise can enhance government-level discussions. This appeal reflected the growing importance of AI across public services, commercial sectors, and infrastructure systems.
Rahman noted that artificial intelligence is already embedded within electricity grids, water systems, health records, economies, and military operations. He warned against an international competition to dominate the technology. Instead, he urged the implementation of practical safeguards capable of mitigating risks while promoting beneficial uses. According to Rahman, these protections should be based on reliable scientific knowledge. His remarks come amid ongoing efforts by governments to establish global frameworks for assessing AI risks, benefits, and governance structures.
UN Develops New Framework for AI Oversight
Over the past year, the United Nations has expanded its formal activities related to artificial intelligence. The General Assembly has established the Global Dialogue on AI Governance along with an Independent International Scientific Panel on Artificial Intelligence. The dialogue’s inaugural session took place in Geneva on July 6 and 7, 2026, bringing together governments, experts, and stakeholders to deliberate on international AI policies. The scientific panel, composed of 40 members, is tasked with evaluating AI opportunities, risks, and broader societal impacts.
Before the start of the General Debate in New York, Rahman also addressed AI governance. On September 10, he discussed how governments and other stakeholders should manage the technologies already transforming societies. Additionally, he met with Doreen Bogdan-Martin, Secretary-General of the International Telecommunication Union, on September 20, to discuss cooperation on AI governance and emerging challenges in advanced AI. They also discussed ongoing collaboration with industry as global efforts in AI regulation evolve.
AI Policy Discussion Part of Broader UN Reform Agenda
Rahman connected artificial intelligence to a wider conversation about the United Nations’ approach to rapidly evolving global issues. He stated that member states and the UN Secretariat remain central to addressing these challenges but also emphasized that outside experts from research, industry, and civil society are increasingly crucial in highly technical areas. Their specialized knowledge can enhance policy discussions. These remarks were made during the General Assembly’s annual high-level meetings in New York.
The 81st session’s General Debate commenced on September 22 and is scheduled to run through September 26, concluding on September 28. Rahman pointed out that the UN has previously faced major technological and security issues, including nuclear proliferation and human cloning, and cited international efforts to protect the ozone layer. Regarding artificial intelligence, he called for government leadership underpinned by scientific and technical expertise. His message underscored the importance of maintaining relevant international governance as AI becomes more prevalent across societies and critical infrastructure systems.
}end{