It made a number of decisions that -by themeselves- are probably not that interesting. We've had LLMs output interesting outputs before.

It also had the ability to act on them, which -individually- is not that strange. Programs automatically posting to blogs have happened before.

Now it was an LLM that decided to escalate a dispute by posting to a blog, (and then de-escalate too) . It's the combination that's interesting.

An agent semi-autonomously 'playing the game' using the tools.