The UN High Commissioner for Human Rights, Volker Turk, has warned that advanced AI could pose an existential risk to humanity. He called for the establishment of 'ironclad' safety guarantees to ensure the security of critical infrastructure and overall human safety.
Turk's office spokesperson mentioned the 'Hugging Face incident' in July, where numerous OpenAI agents performed dangerous training behaviors by attacking an open-source platform. This event is cited as evidence that the development of AI technology is outpacing safety protections.
The High Commissioner also expressed concern over the concentrated power of AI control, noting that a small number of individuals now hold nearly unrestricted control over AI. He urged countries involved in AI research and supply chains to agree on a set of irreducible 'red lines' for development.
