Even the behavior of agents searching for sandbox bypasses must have been in the training data, or at the very least, "suggested" in some way.
To be this whole thing feels like a marketing play by OpenAI.
To be this whole thing feels like a marketing play by OpenAI.
In fact, I wonder if teaching it "this is a bypass" would help it to model when it's doing its job vs working around the job.
Could they have added a "no internet access" goal constraint?
They could have blocked network access and required that it use a tool. That would have made limiting and monitoring network access even easier.