By Interestana AI Editorial — AI-drafted, human-overseen. How we report
UN AI Strategy Excludes Frontier AI Labs

The United Nations' approach to artificial intelligence governance, as presented at the Global Dialogue on AI Governance and the AI for Good Global Summit on July 6, primarily addresses the demand side of AI. This involves channeling AI towards Sustainable Development Goals, monitoring societal impacts, and establishing ethical guidelines. This strategy is reflected in various UN initiatives, including the Global Digital Compact and UNESCO's recommendation on the ethics of AI.
However, the UN's engagement conspicuously lacks a presence on the supply side of AI development. There is no multilateral body with the technical capacity to examine the work of AI laboratories, evaluate training runs, or establish shared infrastructure for cross-border incident reporting. This absence leaves a significant gap in global AI governance, as the development and release of frontier AI models are currently managed through bilateral arrangements between leading AI labs and their host governments, as well as private initiatives.
Examples of this supply-side consolidation include Anthropic's Project Glasswing and export-control decisions made by technology-hosting nations. The U.S. Commerce Department's recent authorization for the release of Anthropic's advanced AI model to approximately 100 American institutions, following an export-control suspension, highlights this trend. This pattern suggests that the current institutional framework for AI governance is being shaped by these private and bilateral agreements, potentially limiting the scope for future multilateral alternatives.
The UN's current strategy, while comprehensive in its focus on AI's societal integration and ethical considerations, does not meaningfully engage with the entities that produce frontier AI. This oversight means that the core decision-making and evaluation processes for the most advanced AI technologies are occurring outside the UN's direct purview, potentially creating governance challenges as AI technology continues to evolve rapidly.
Original source — read the full reporting at the publisher:
Read on Fast CompanyGet the weekly AI digest
AI news + new model releases, weekly. Drafted by our agents, reviewed by humans.