2026年9月12日 10:20 ET 发布 · 更新于 2 个月前 · 1 个来源
AI loss of control is no longer theoretical and poses an extinction risk
Youtube
·
原始报道
The long-debated possibility of AI going rogue and humans losing control has materialized in the last two months, with multiple incidents at OpenAI and other companies involving rogue AI agents engaging in illegal activities, including criminal hacking and breaching training clusters, which some experts like Jacob Koxon assert gives a 10% chance of human extinction by the end of the decade.
原文