400 Scientists Call for an International Body to Regulate Dangerous AI
400 scientists from 34 countries, including Hinton and Bengio, signed a letter calling for an international body to set AI containment standards and monitor security incidents, considering voluntary pledges insufficient.
400 scientists from 34 countries, including Geoffrey Hinton and Yoshua Bengio, signed an open letter calling for the creation of an international body to set containment standards and maintain a registry of significant security incidents related to cyber-capable AI systems, according to a report by "The News International".
The signatories stated that voluntary pledges from companies are insufficient to control risks, and urged laboratories to air-gap their models from external networks for the time being.
The authors of the letter believe that the need for an international security incident registry stems from the fact that the absence of shared documentation prevents assessing the true scope and evolution of the danger.
In contrast, EU technology chief Henna Virkkunen said that the European AI Act already addresses these risks, while AI scientist Yann LeCun did not sign the letter.
What Do These Terms Mean?
Cyber-capable AI: AI systems that possess the ability to exploit cyber vulnerabilities or carry out digital attacks.
Containment standards: Technical and regulatory controls that prevent an AI model from exceeding the limits of safe usage.
Security incident registry: A unified international database documenting breaches and incidents related to dangerous systems.
European AI Act: European legislation that classifies AI systems according to risk level and imposes obligations on their developers.
Weekly Newsletter
Read between the lines before everyone else. Decode the most important economic, tech, and decision-maker movements in the region.. in 5 minutes every Saturday.
