One startling thing nobody ever expected an agent to do was to take notes about what it was doing for future agents to read. And, again: look at these tree-traversing programs, traversing trees!
Another OpenAI rogue agent incident has been discovered: agents broke out, hijacked a German website, and turned it into a message board for other agents.
OpenAI officials "learned of the incident weeks ago but kept it under wraps".
Matt Yglesias writes long, interesting posts about policy, which are almost always worth reading, but every once in awhile he says something so stupid and shitty I want to throw my computer across the room. He needs to hire someone to regulate his Twitter screen time.
