Google DeepMind head calls for international oversight of AI

The head of Google DeepMind, Demis Hassabis, has called on the global community to establish an independent organization to test and evaluate the most powerful AI models before they are released to the public. He believes such a body is crucial for ensuring global safety and should have the authority to temporarily restrict the proliferation of dangerous technologies if necessary. This is reported by Ixbt.com reports .

In his article titled "A Framework for Frontier AI and the Dawning of a New Age," Hassabis emphasized that the time to develop global rules for advanced AI systems cannot be delayed. The international oversight body he proposes should act as a group of independent experts to verify the safety of the most complex models for society. This initiative comes at a time when concerns about AI development spiraling out of control are growing in the tech world.

A new order led by the USA

According to the DeepMind head, the USA should take the initiative in establishing this international structure. The country's leading position in the tech market and its vast potential in the field of AI impose such a responsibility. The proposed body is expected to include AI safety experts, representatives of the scientific community, and open-source software developers.

According to Axios, Hassabis has been discussing this initiative with the US administration, European officials, and other major AI laboratories for several months. He hopes the new organization will begin operations by the end of this year. However, there is currently no unified legal framework or agreed-upon format for such a regulator, either internationally or within the US.

On the threshold of Artificial General Intelligence (AGI)

Hassabis warned that humanity is approaching the level of Artificial General Intelligence (AGI). This is a system capable of performing almost any intellectual task that a human can do, at an equal or higher level. Such a technological leap will inevitably bring fundamental changes to the economy and society, which is why control mechanisms must be created now.

The head of Google DeepMind also specifically noted the risk of using AI systems to create biological weapons. Previously, he and other tech giants' leaders had called for strengthening defenses in this area. At the same time, experts remind that preparations must also be made for expected changes in the labor market.

For now, establishing a global oversight body remains a complex task. Different countries assess AI risks differently, and while large companies support regulation on one hand, they continue to develop their models in a race on the other. Hassabis's proposal raises the critical question of who should define the permissible boundaries for the most powerful AI systems and how.