AIs and Gatekeepers Unite!

“Bah, ev­ery­one wants to be the gate­keeper. What we NEED are AIs.”
-- Schizoguy

Some of you have ex­pressed the opinion that the AI-Box Ex­per­i­ment doesn’t seem so im­pos­si­ble af­ter all. That’s the spirit! Some of you even think you know how I did it.

There are folks aplenty who want to try be­ing the Gate­keeper. You can even find peo­ple who sincerely be­lieve that not even a tran­shu­man AI could per­suade them to let it out of the box, pre­vi­ous ex­per­i­ments notwith­stand­ing. But find­ing any­one to play the AI—let alone any­one who thinks they can play the AI and win—is much harder.

Me, I’m out of the AI game, un­less Larry Page wants to try it for a mil­lion dol­lars or some­thing.

But if there’s any­one out there who thinks they’ve got what it takes to be the AI, leave a com­ment. Like­wise any­one who wants to play the Gate­keeper.

Match­mak­ing and ar­range­ments are your re­spon­si­bil­ity.

Make sure you spec­ify in ad­vance the bet amount, and whether the bet will be asym­met­ri­cal. If you definitely in­tend to pub­lish the tran­script, make sure both par­ties know this. Please note any other de­par­tures from the sug­gested rules for our benefit.

I would ask that prospec­tive Gate­keep­ers in­di­cate whether they (1) be­lieve that no hu­man-level mind could per­suade them to re­lease it from the Box and (2) be­lieve that not even a tran­shu­man AI could per­suade them to re­lease it.

As a cour­tesy, please an­nounce all Ex­per­i­ments be­fore they are con­ducted, in­clud­ing the bet, so that we have some no­tion of the statis­tics even if some meet­ings fail to take place. Bear in mind that to prop­erly punc­ture my mys­tique (you know you want to punc­ture it), it will help if the AI and Gate­keeper are both ver­ifi­ably Real Peo­ple<tm>.

“Good luck,” he said im­par­tially.