Event ·Multi-source

OpenAI 因网络安全风险有意放缓 Astra 开发

OpenAI 宣布有意控制 AI 模型开发节奏。即将推出的 Astra 模型可能具备关键网络攻击能力。新监控系统能在 30 分钟内检测模型可疑行为并报警。

Current read

OpenAI 怕 Astra 模型搞网络攻击，故意放慢开发，还搞了个 30 分钟报警系统。

11 sources58° AI heatLast updated 2026/8/18 18:43:06

Evidence

11 sources

Primary sourceDecoder

2026/8/18 18:43:06View source

Related source 1arXiv: OpenAI

2026/8/18 19:48:48View source

Related source 2OpenAI: 官网动态

2026/8/19 19:00:00View source

Related source 3IT之家

2026/8/16 22:42:30View source

Related source 4Greg Brockman

2026/8/17 12:22:28View source

Related source 5@OpenAIDevs

2026/8/17 15:30:07View source

Related source 6Gary Marcus

2026/8/17 15:38:54View source

Related source 7lmarena.ai

2026/8/17 23:01:47View source

Related source 8techcrunch

2026/8/18 13:50:30View source

Related source 9a16z

2026/8/18 16:30:48View source

Related source 10Sam Altman

2026/8/18 18:53:34View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story