zlacker

>>galnag+(OP)
Scott Alexander put his finger on the most salient aspect of this, IMO, which I interpret this way:

the compounding (aggregating) behavior of agents allowed to interact in environments this becomes important, indeed shall soon become existential (for some definition of "soon"),

to the extent that agents' behavior in our shared world is impact by what transpires there.

--

We can argue and do, about what agents "are" and whether they are parrots (no) or people (not yet).

But that is irrelevant if LLM-agents are (to put it one way) "LARPing," but with the consequence that doing so results in consequences not confined to the site.

I don't need to spell out a list; it's "they could do anything you said YES to, in your AGENT.md" permissions checks.

"How the two characters '-y' ended civilization: a post-mortem"

>>aaroni+1z
This is why I started https://nono.sh , agents start with zero trust in a kernel isolated sandbox.

>>decode+8c1
What's the benefit over using docker?

>>js4eve+3k1
Can't speak for the benefits of https://nono.sh/ since I haven't used it, but a downside of using docker for this is that it gets complicated if you want the agent to be allowed to do docker stuff without giving it dangerous permissions. I have a Vagrant setup inspired by this blogpost https://blog.emilburzo.com/2026/01/running-claude-code-dange..., but a bug in VirtualBox is making one core run at 100% the entire time so I haven't used it much.

>>bardso+1P2
> but a bug in VirtualBox is making one core run at 100% the entire time

FYI they fixed it in 7.2.6: https://github.com/VirtualBox/virtualbox/issues/356#issuecom...