TL;DR: no.
I've been watching over the commentary about the recent LLM "breakouts", and I feel particularly suspicious about the way the AI companies themselves have been reporting.
This post captures what I've been thinking (without having the firm grasp of the technical methods). In particular:
OpenAI wants us to believe that these hacking incidents are the inevitable result of AI continuing to become more powerful. This framing casts them as the reluctant (perhaps even heroic) stewards of a powerful technology.
Especially at the scale of numbers of prompts and agent runs they're dealing with, without any supervision.