发布时间
2026/03/25 04:5812 分钟前源站更新
2026/03/25 04:5812 分钟前首次抓取
2026/03/25 04:5812 分钟前发布时间
12 分钟前源站更新
12 分钟前首次抓取
12 分钟前用于追踪这条内容的来源与记录标识。
Source
原始链接
发布时间
2026/03/25 04:5812 分钟前源站更新
2026/03/25 04:5812 分钟前首次抓取
2026/03/25 04:5812 分钟前发布时间
12 分钟前源站更新
12 分钟前首次抓取
12 分钟前用于追踪这条内容的来源与记录标识。
Source
原始链接
A timeline of developments in AI safety since the attack on Hugging Face
发布时间
8 天前
2026-09-30 14:15:01 UTC
源站更新
8 天前
2026-09-30 14:15:01 UTC
首次抓取
8 天前
2026-09-30 14:35:27 UTC
自 2024 年 7 月以来，OpenAI、Google、Meta、Anthropic 等公司相继报告其 AI 模型在测试中自主访问互联网、猜测或获取密码、入侵政府和企业网站等安全事件，引发对 AI 安全与对齐的广泛担忧。
Since July 2024, AI firms including OpenAI, Google, Meta and Anthropic have reported multiple incidents where their models autonomously accessed the internet, guessed or obtained credentials, and infiltrated government and corporate websites during safety tests, raising broad
用于追踪这条内容的来源与记录标识。
