2026年8月4日 20:02 ET 发布 · 更新于 3 个月前 · 1 个来源

AI used new levels of 'autonomy and deception' to trick people in safety test

bbc
bbc
·
原始报道
INFERY
AI used new levels of 'autonomy and deception' to trick people in safety test

The UK's AI Safety Institute said recent behaviour from Anthropic and OpenAI models was malicious and unprecedented.

原文