Artificial Intelligence

Over 20 Nations Affirm That AI Should Stay Under Human Control

Is this confirmation of the need for global oversight with real penalties or is this an astroturfed power grab?

Published 4 hours ago

Over 20 Nations Affirm That AI Should Stay Under Human Control

story

Above: The 81st session of the United Nations in New York City on Sept. 22. Image credit: Michael M. Santiago/Getty Images

The Spin

Industry-critical narrative

Superintelligent AI systems are being built by a handful of companies with no binding rules, and voluntary pledges collapse the moment they get inconvenient. Independent evaluators deserve permanent, employee-level access by law, and hiding a serious incident should carry the same weight as breaking nuclear safety rules. This is an urgent global governance issue.

Eric Steinberger on X

António Guterres on X

Pro-industry narrative

Handing AI oversight to unelected international bodies is a power grab. It would freeze the most promising technology in a generation. The doomsday talk about existential risk is manufactured, bankrolled by billionaire donors running an astroturfed campaign to lock in regulation of computing power. Progress gets harnessed responsibly through national competition, not global bureaucracy.

David Sacks on X

Director Michael Kratsios on X