A few months ago when I started yoloAI, it was interesting to watch the resistance to the very idea of sandboxing an AI agent. Even with a few high profile examples already out there of misguided agents wreaking havoc, the consensus persisted: Everybody knows “We don’t need no stinkin’ sandboxes!” The built-in agent controls are more than enough! yoloAI was one of very few projects out there arguing otherwise, and the github stars (a poor gauge of utility, I know) remained elusive.
Fast forward a few months, and just this past week I noticed a shift in consensus. Today, everyone knows (and always has known, apparently) that “of course agents should be sandboxed! I mean, that’s just basic due diligence!”
But it comes with a twist: Everybody also knows that AI sandboxing is easy. Just spin up a VM, seed it with what you need, maybe add some firewall rules, and you’re off to the races!
Except it’s not easy. Not by a long shot.
I look forward to the day where everybody finally knows: “Well of course sandboxing AI agents is difficult! DON’T ROLL YOUR OWN! You should use a sandboxing technology built by an expert if you value your security and infrastructure!”
We’re not there yet.