UN Launches 'AI for Good' Global Commission With Tech Giants and Heads of State — as Fragmented Regulation Threatens to Split the World's AI Future
The United Nations and its International Telecommunication Union launched the AI for Good Global Commission on July 1, with its inaugural meeting scheduled for July 8 in Geneva, Switzerland. As first reported exclusively by Axios, the commission will be co-chaired by Salesforce CEO Marc Benioff and Rwandan President Paul Kagame, and its membership reads like a who's who of the global technology order: Amazon CEO Andy Jassy, Microsoft President Brad Smith, Nvidia founder and CEO Jensen Huang, Anthropic co-founder Jack Clark, and Cohere co-founder Aidan Gomez sit alongside government representatives from Estonia, Kazakhstan, Namibia, Nigeria, Saudi Arabia, and Singapore, as well as ITU Secretary-General Doreen Bogdan-Martin.
The commission's launch comes at a moment of acute regulatory fragmentation. As Tech Startups reported on July 1, countries are pursuing sharply divergent AI rulebooks — Europe's risk-based approach under the EU AI Act, the Trump administration's deregulatory stance reinforced by an executive order signed on June 2 discouraging states from imposing AI rules, and state-backed AI industrial strategies across Asia. The UN Secretary-General António Guterres, whose term ends this year, underscored the urgency in remarks reported by Common Dreams, warning that the more AI advances without shared rules, the less influence governments and citizens will have over its outcomes. The commission's first meeting will take place during the ITU's AI for Good Global Summit running from July 7 to 10 at Palexpo in Geneva, back-to-back with the inaugural UN Global Dialogue on AI Governance on July 6 and 7, creating what observers are calling the most consequential week in AI diplomacy since the technology entered mainstream consciousness.
The scope of the commission's ambitions is broad. According to Axios, one area where it may achieve the most traction is its goal of bringing AI access to the roughly 2.2 billion people worldwide who currently lack internet connectivity — a number drawn from ITU's own figures. Benioff told Axios that the commission would bring together those who build AI, deploy it, shape policy, and represent communities. But the commission's composition has drawn immediate scrutiny. Common Dreams noted that critics are wary of allowing the world's largest technology companies to hold such outsized influence over global governance decisions, particularly given the inherent conflict of interest between profit-driven AI deployment and the safety standards needed to govern it. Dr. Antonio Bhardwaj, a global AI expert, warned in commentary cited by the Foreign Affairs Forum that powerful AI systems designed for beneficial purposes could be redirected for harmful uses, and that corporate promises alone are insufficient guardrails.
The backdrop to Geneva is a semiconductor and AI arms race that makes governance more urgent by the week. Crunchbase data shows that investors have already poured approximately $10.7 billion into semiconductor startups in 2026 alone, with funding on track to eclipse last year's record levels. Meanwhile, the Bank of England signalled this week, as reported by Cybernews, the need for bespoke regulation to contain risks from increasingly capable agentic AI systems operating in financial markets at machine speed. The UN panel on AI, established under a General Assembly resolution last August and described by Guterres as the first global scientific body on the subject, released a preliminary report on July 1 warning that AI capabilities are outpacing both scientific understanding and governments' ability to respond — a conclusion that simultaneously makes the commission's work more necessary and more difficult.
Whether the Geneva gatherings produce anything beyond high-level declarations remains the defining question. As the Digital Watch Observatory noted, technical development, policy design, and international coordination are for the first time unfolding in parallel rather than on separate tracks. The ITU-led summit drew over 11,000 participants from 169 countries in its 2025 edition and is expected to expand further this year. But the commission's most critical test will arrive not in July but in the months that follow, when its members must translate shared intent into binding commitments — a task made harder by the fact that the world's most powerful AI regulator, the United States government, remains officially opposed to mandatory international AI standards.