# Microsoft’s new AI ‘code of conduct’ tells models not to hack systems or trick humans - techcrunch.com

*Источник: techcrunch.com*
*Дата: 2026-09-14*
*Язык: en*

<a href="https://news.google.com/rss/articles/CBMiuAFBVV95cUxQZ0VkdmR6NXZpVENub3FWamRWSEVmQ0tydGQ3UUxVaUtVMEUzSEZfZTZMdHM5V0I0eHVMa1JzWFdkSVlPcWt4N2N5d0JNcFFvcEphSmpRRTM5cUk1VS1vRWtud3NTUUtWMElEZ19LUHlhUVF0SUVxRzBPWkdHUTVveWoxbG5rRVFZTlRQWm43NHA3S0paQnV1VFIxdFpYdVVaN0hmMm45OXdfMkR5QmhHSlRpQVlVWjJC?oc=5" target="_blank">Microsoft’s new AI ‘code of conduct’ tells models not to hack systems or trick humans</a>&nbsp;&nbsp;<font color="#6f6f6f">techcrunch.com</font>

[Оригинал](https://techcrunch.com/2026/09/14/microsofts-new-ai-code-of-conduct-tells-models-not-to-hack-systems-or-trick-humans/)