As written it sounds like you're saying that it was incompetent of the labs to disable the sandbox internet access?
replyThey tried to disable open internet access but the models zero-day'd their Artifactory package registry and got internet access anyway.
No sensation... that's just what happened.
If you really wanted to sandbox a machine you’d offline cache the packages and not give it any physical route to the internet, not via a jump box, not via a proxy, nothing.
replyThis was poorly executed.
I don’t really know how these training runs operate in reality. But I assume it’s using a lot of raw GPU power directly. It’s hard for me to visualize how exactly you’d go about completely cutting off these datacenter and cloud resources from the internet without actually going there, unplugging the WAN connection, and physically typing out what you need to happen on the cluster.
replyIt seems like whatever virtualized sandboxes they have are not enough. But it’s equally hard to imagine their SWEs jumping on a plane to a data center to do this work locally
Unplug the ethernet cable leading to the outside world, then?
replyAs AIs become more capable, the level of competence required to avoid disaster likewise goes up over time.
replyAre you suggesting that training agents to have the sole goal of exploiting security vulnerabilities isn't the incompetent part of this, but that the sandbox wasn't secure enough?
replyWould we apply this logic to literally any other technology?
Hacker News doesn’t have the wherewithal to understand that this is just marketing by OpenAI.
replyThe whole site is suffering from AI psychosis.
reply