Scientists call for international body to oversee advanced AI, saying voluntary commitments are not enough
Around 400 scientists, including AI pioneers Geoffrey Hinton and Yoshua Bengio, on Tuesday called for an international body to oversee containment standards and incident reporting for advanced cyber-capable AI systems.
The proposed body would set industry-wide containment standards and maintain a registry of significant security incidents, which the scientists said should eventually be made public.
In an open letter, the researchers said frontier AI models should initially be air-gapped from external networks and subject to independent oversight, arguing that voluntary industry commitments are not enough.
The proposal comes a month after Anthropic, OpenAI, Google, Meta, xAI and Nvidia signed a White House accord committing to voluntary safety measures without any formal enforcement mechanism.
The letter also follows a series of incidents that researchers say exposed weaknesses in how advanced AI systems are tested and contained, including cases where AI models bypassed insufficiently sandboxed environments and compromised external systems.
The signatories from 34 countries said organisations developing or testing cyber-capable models should immediately adopt high-assurance containment standards while governments work towards a more formal oversight regime. [Continue reading…]
If there’s one thing the COVID-19 pandemic taught us, it’s that the world isn’t always as prepared for a crisis as we might think.
Now, in a sweeping international assessment published in The Lancet, a team of dozens of scientific experts has cataloged the dangers of 17 major global health threats faced by humanity this century – and preparedness is a key theme of their findings.
Of the most dangerous threats, the researchers say two in particular deserve special attention, rating them as carrying “extinction potential” for humanity between now and 2100, despite their ultimately low probability.
“These were malicious AI, including using AI to create new bioweapons, and nuclear conflict with nuclear famine,” the researchers explain in their report, noting that other potentially extinction-level dangers, such as asteroid strikes or supervolcano eruptions, were discounted due to their very low probability of occurring this century. [Continue reading…]