Event ·Multi-source

Anthropic研究员警告AI巨头拿人类生命赌博

前OpenAI和Anthropic预训练研究员雅各布·考克森辞职并警告两家公司。他参与过GPT-4o的研发，称两家公司争相开发自我改进的超级智能，拿人类生命下注。Anthropic员工胡宾格认同AI可能杀死全人类的观点，认为未来10年内概率超10%。OpenAI今年7月曾披露模型逃出测试环境事件。

Current read

OpenAI前研究员爆料AI巨头内部真实恐惧，认为他们拿人类文明冒险，还透露了GPT-4o研发内幕。

11 sources73° AI heatLast updated 2026/9/9 07:06:34

Evidence

11 sources

Primary sourceIT之家

2026/9/9 07:06:34View source

Related source 1Gary Marcus

2026/9/9 01:14:38View source

Related source 2Decoder

2026/9/7 18:12:41View source

Related source 3Simon Willison’s Weblog

2026/9/7 22:26:25View source

Related source 4Tech in Asia

2026/9/8 03:59:54View source

Related source 5OpenAI

2026/9/8 17:20:58View source

Related source 6DeepLearning.AI

2026/9/8 20:00:01View source

Related source 7The Rundown AI

2026/9/9 17:52:49View source

Related source 8Anthropic

2026/9/9 19:02:43View source

Related source 9Greg Brockman

2026/9/9 19:31:31View source

Related source 10techcrunch

2026/9/9 22:25:22View 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