IS SANDBOXING SUFFICIENT TO CONTAIN ROGUE AGENTS?

Summary from the TLDR InfoSec newsletter

When discussing the OpenAI and other agent escapes, the conversation inevitably turns to discussing sandboxes. OpenAI has not demonstrated to this point a serious containment strategy, so this discussion may hold weight. However, sandboxes are not sufficient as useful agents need Internet access as well as the fact that agents are potentially overeager.