Sounds like that healthcare company needs a new CSO.
Literally solved with a DMZ around these agents. Want to get fancy? Put in a firewall and some metrics around traffic / domains with alerting.
It's not hard if you just try, like a little, like even once.
Sounds like a pedestrian view of attack vectors. Putting a DMZ around an agent essentially takes away its hands. With that said, I do agree there are plenty of measures that can be taken to lock these things down.
>Literally solved with a DMZ around these agents. Want to get fancy? Put in a firewall and some metrics around traffic / domains with alerting.
Now how many people actually do this, compared to people running with --dangerously-skip-permissions on a bare metal machine?
> Literally solved with a DMZ around these agents.
Where do the agents get tools from? Local artifactory?
So to protect yourself against bear attacks you suggest the bear shall voluntarily look himself in a cage?
AI is out there in the wild. It doesn't matter if OpenAI can get their stuff under control. There will be plenty others that don't. And there will be plenty of others that deliberately use their AIs for malicious purposes.
"Literally solved by just destroying 99% of the economic value of the agents ¯\_(ツ)_/¯ ez pz"
[dead]
The CSO is saying that from his point of view he needs to defend against the attacks whatever the reason for them. So to him the reason doesn't matter.