The UN Just Created an IPCC for AI—And Only Two Countries Objected

When the Intergovernmental Panel on Climate Change released its first assessment report in 1990, few could have predicted how profoundly it would reshape global policy. For more than three decades, the IPCC has served as the definitive scientific voice on climate change, its findings guiding everything from international treaties to national energy policies. Now, the United Nations is betting that artificial intelligence deserves the same treatment.

“The panel will act as an early-warning system and evidence engine, helping distinguish between hype and reality.” — United Nations statement

A $25 Million Bet on Evidence

On February 12, 2026, the UN General Assembly voted to establish the Independent International Scientific Panel on Artificial Intelligence. The 40-member body, drawn from 37 nations, represents one of the most ambitious attempts yet to create a credible, global authority on AI’s risks and opportunities. Unlike previous efforts that focused narrowly on safety or ethics, this panel’s mandate is sweeping: economic impacts, social consequences, cultural shifts, and developmental implications are all fair game.

The comparison to the IPCC is intentional and, according to observers, apt. “The new panel will produce a number of scientific reports, year on year, which will be broad ranging—not just safety,” says Wendy Hall, a computer scientist at the University of Southampton who served on the panel’s precursor body. Brian Tse, founder of Beijing-based AI safety consultancy Concordia AI, notes that the group is “much bigger in scope and is truly global.”

Who’s On the Panel

Yoshua Bengio, the Montreal-based deep learning pioneer and vocal AI safety advocate, brings academic credibility. Balaraman Ravindran of IIT Madras represents India’s growing AI research community. Jian Wang, founder of Alibaba’s cloud computing arm, provides industry perspective. Nobel Peace Prize laureate Maria Ressa, the Manila-based journalist, adds a voice attuned to information integrity and democratic values.

At least nine members have industry backgrounds, though all serve in personal capacities. The selection process, led by three UN technical agencies, reviewed more than 2,600 candidates—a level of competition that suggests serious intent.

“The panel’s broad geographical and cultural diversity will be one of its greatest strengths. The impacts of AI are highly context dependent.” — Brian Tse, Concordia AI

The Two Dissenters

Not everyone embraced the initiative. The United States and Paraguay voted against the panel’s formation, making them the only two nations to formally object. The US opposition is particularly notable given its dominant position in AI development—American companies and research institutions produce much of the technology the panel will study.

The dissent raises questions about whether the panel can achieve the consensus-driven influence of its climate counterpart. The IPCC’s power derives partly from its ability to speak with one voice, even when governments disagree. If major AI powers remain skeptical, the new panel’s reports risk being dismissed as politically motivated rather than scientifically grounded.

What the Panel Won’t Do

Importantly, the panel has no regulatory authority. It cannot set policy, issue binding standards, or impose penalties. Its power lies entirely in its ability to synthesize research and communicate findings clearly—a model that worked for climate science but faces different challenges with AI.

Climate change, for all its complexity, is fundamentally a physical phenomenon. AI’s impacts are social, economic, and cultural—domains where scientific consensus is harder to achieve and political disagreement runs deeper. The panel’s success will depend on whether it can navigate these murkier waters without losing credibility.

The first reports are expected within the year. Until then, the AI world will watch to see whether this new body can fulfill its promise—or whether it becomes another well-intentioned but ultimately ignored forum in an already crowded landscape of AI governance initiatives.

This article was reported by the ArtificialDaily editorial team. For more information, visit Nature and United Nations.