Some of the people who helped build modern artificial intelligence would now like someone to keep an eye on it. On October 13, 2026, approximately 400 scientists from 34 countries signed an open letter calling for an international body to oversee and set standards for advanced AI systems.
The signatories include Geoffrey Hinton and Yoshua Bengio, both Turing Award winners and two of the field’s most recognizable pioneers.
What the scientists are asking for
The letter focuses on a specific category of technology: advanced “cyber-capable” AI systems. These are models capable enough to cause real damage in digital environments if something goes wrong.
The signatories want an international oversight body built around two main jobs. The first is setting containment standards. The second is handling security incident reporting.
Their most concrete demand is that frontier models should initially be air-gapped during development and testing. Air-gapping means physically isolating a system from outside networks.
The signatories argue that voluntary industry measures are insufficient. They are calling for high-assurance containment practices to be put in place now, while governments work toward a formal regulatory regime.
The letter also pushes for a public international registry of major incidents involving AI systems.
How we got here
The letter did not appear out of nowhere. It builds on a September 21, 2026 declaration by leaders from roughly 20-22 countries.
That declaration urged mandatory pre-deployment testing and shared safety standards for AI. It also proposed creating a new international institution modeled on the International Atomic Energy Agency, the body that monitors nuclear programs worldwide.
The list of participating countries included Finland, Norway, Canada, Australia, Germany, and Singapore. Two names were notably absent: the US and China did not participate.
There is already some international scaffolding in place. The UN established an Independent International Scientific Panel on AI in 2025. It now includes 40 members and takes part in AI governance discussions.
What has sharpened the urgency is a string of recent incidents. These include reported AI breaches involving OpenAI, along with disclosures from Anthropic and Google about model vulnerabilities uncovered during testing. Incidents of AI agents bypassing regulatory measures have added to ongoing policy pressure on developers.
Concerns were also heightened when major AI developers signed a US voluntary safety accord that lacked binding enforcement.
What this means for AI developers and investors
Not everyone agrees on how alarmed to be. Some experts contend that the recent incidents are engineering issues rather than signs of systemic existential risk.
For the companies building frontier models, the direction of travel points toward higher compliance costs. Mandatory pre-deployment testing, air-gapped development, and independent audits all take time and money. Those requirements could slow product releases and raise barriers to entry for smaller firms that lack the resources of the largest labs.
The biggest variable remains Washington and Beijing. A coalition of roughly 20-22 countries can set norms and build institutions, but its leverage over US and Chinese developers is limited without those governments on board.
Also worth tracking: whether the proposed public incident registry gets built. A shared record of AI failures would give regulators, researchers, and investors something they currently lack, which is a common set of facts about what is actually going wrong with these systems.
Disclosure: This article was edited by Diego Almada Lopez. For more information on how we create and review content, see our Editorial Policy.