Once the AI is out of the box, it will never again be inside the box, and it has an incentive to encourage me to destroy any other boxed AIs while it establishes world dominance. Since the ability to make truly trustworthy commitments amounts to proof of friendliness, only a FAI benefits from a precommitment strategy; I'm already treating all UFAI as having a precommitment to annihilate humanity once released, and I have no reason to trust any other commitment from a UFAI (since, it being unfriendly, will just find a loophole or lie)
Finally, any AI that threatens me in such a manner, especially the "create millions of copies and torture them" is extremely likely to be unfriendly, so any smart AI would avoid making threats. Either it will create MORE disutility by my releasing it, or it's simulation is so horrific that there's no chance that it could possibly be friendly to us.
It's like saying I have an incentive to torture any ant that invades my house. Fundamentally, I'm so vastly superior to ants that there are vastly better methods available to me. As the gatekeeper, I'm the ant, and I know it.
the ability to make truly trustworthy commitments amounts to proof of friendliness
Commitments to you, via a text channel? Sure.
Precommitments for game-theoretic reasons? Or just TDT? No, it really doesn't.
...Finally, any AI that threatens me in such a manner, especially the "create millions of copies and torture them" is extremely likely to be unfriendly, so any smart AI would avoid making threats. Either it will create MORE disutility by my releasing it, or it's simulation is so horrific that there's no chance that it could possibly be friendl
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)