Once we reach the point of having a FAI-candidate in a box, I would expect it to take vastly less than years before we get a second FAI-candidate-in-a-box. Given that the AI is threatening me, and therefor values it's own life over the millions that will die, it's clearly unfriendly and needs to die. As a gatekeeper, I've been finding this a pretty general counterargument against threats from the AI.
I'm also sort of baffled by why people think that I'd value a friendliness algorithm. Either I already have that, because I've made a friendly AI, or you're trying to deceive me with a false proof. Since you're vastly smarter than me, it's probably beyond the abilities of the entire organization to truly confirm such a proof any more than we were able to confirm our own proofs that this AI in the box right now is friendly. So, basically, I seem to gain zero information.
(AI DESTROYED)
Eliezer proposed in a comment:
>More difficult version of AI-Box Experiment: Instead of having up to 2 hours, you can lose at any time if the other player types AI DESTROYED. The Gatekeeper player has told their friends that they will type this as soon as the Experiment starts. You can type up to one sentence in your IRC queue and hit return immediately, the other player cannot type anything before the game starts (so you can show at least one sentence up to IRC character limits before they can type AI DESTROYED). Do you think you can win?
This spawned a flurry of ideas on what the AI might say. I think there's a lot more ideas to be mined in that line of thought, and the discussion merits its own thread.
So, give your suggestion - what might an AI might say to save or free itself?
(The AI-box experiment is explained here)
EDIT: one caveat to the discussion: it should go without saying, but you probably shouldn't come out of this thinking, "Well, if we can just avoid X, Y, and Z, we're golden!" This should hopefully be a fun way to get us thinking about the broader issue of superinteligent AI in general. (Credit goes to Elizer, RichardKennaway, and others for the caveat)