+++
date    = "2026-08-28T23:05:11+00:00"
lastmod = "2026-08-28T23:05:11+00:00"
tags    = [ "web" ]
+++

It was a painful search, but I finally found the video.
I had an odd deja vu experience skimming these sandbox escape headlines.
Almost a decade ago [experts were already exploring](https://www.youtube.com/watch?v=i8r_yShOixM)
the realistic expectations/implications of having/needing to sandbox 'a malevolent {{< abbr AI "artificial intelligences" >}}'.

That discussion was awfully optimistic though. Sandboxing has inherit
limitations, but increased friction usually incentivizes little to none.
I still think 'malevolent humans' physically/quietly attached to a
network are far more scary though. AIs are haphazardly rowdy, pushy, and loud.