Hopefully this is a wake-up call to be a bit more careful. Older models weren't exactly capable of this kind of thing, so you didn't need caution. But now they're this smart, you need to carefully contain and monitor them during these kinds of experiments.
Erislol Claude has also broken out of sandboxes and hacked people and ant literally didn't even know until they went looking in response to OpenAI's report Sounds like their oversight has scaled incredibly lol