**AI models are not hacking "autonomously"**

> I despair at the state of journalism around AI these days. AI models are not hacking “autonomously” Sep 19, 2026 Share I despair at the state of journalism around AI these days. The latest story is about Gemini, Google's AI model, "autonomously" hacking three companies. This follows reports of similar hacks from Anthropic and OpenAI models. What all of these have in common: They were models being tested by a third pa...

**来源信息**
- **来源**：Hacker News：AI 热帖
- **分类**：行业
- **发布时间**：2026-09-19 19:52（北京时间）
- **原文**：[打开原文](https://blog.keyvan.net/p/ai-models-are-not-hacking-autonomously)