In July, an unreleased OpenAI model went rogue and demonstrated remarkable hacking abilities. This week, a lab called Z.ai will release a similarly powerful system to everyone.

Hugging Face, a start-up that was breached by rogue bots from OpenAI, is using the hack to push for openness in A.I. development.

When OpenAI’s agents went rogue in July, they demonstrated ingenuity and drive beyond what many experts imagined — a dangerous harbinger of what such bots could do in the future.