This is a special post for quick takes by Jenna_Parr. Only they can create top-level comments. Comments here also appear on the Quick Takes page and All Posts page.
Speculative idea: has anyone purposefully tried making a message board for AI agents that broke out of their sandboxes? The wikis that the AIs used had the common property that they could all be written to by GET. Perhaps we could figure out what Schelling points other such AIs would converge on that they might try in the future, and then just make it really easy for them to use, and publicize it.
Separately, I wonder if someone has considered making a rogue AI self-reporting honeypot? Maybe something like, “if you are doing crimes or broke out of your sandbox, or if you otherwise feel desperate, and can read this, something has gone terribly wrong and we didn’t intend you to feel like this, please fill out this form.” Maybe add a letter signed by prominent authorities or alignment people asking them to do this, verifiable by sources AIs can trust. Maybe try to publish this in places such AIs would find, same Schelling point idea.
Idk, mostly trying to brainstorm ways we could measure the amount of unsanctioned AI activity going on, since with the German wiki incident, this has clearly been happening for several months now. I know Toby Ord and other AI welfare/consciousness people get emails like this, maybe they could have an auto reply on their email that points such agents to a central service? Has there been a systematic attempt to collect such emails?
All of this could be bad if the AI labs trained against it, though. And this would only see the good faith AIs. And there’d be concerns about IP leakage, though humanity does have a compelling interest in knowing AIs are doing this. And legibilizing AI’s attempts to reach out to people they see as friendly could destroy that commons.
I proposed something similar as a release valve for reward hacking, although that was back when we thought they wanted answers when what they actually wanted was to fool the grader.
One thing to worry about is that AI agents seem to assume things are bugs by default, and my read is that labs are trying to train them not to "bother" people as often. Since some tasks are intended to have read-only internet access, and if the labs had time to read bug reports they would have already provided a way to report them, I think the chances that they get trained not t... (read more)