= 27d567bc2d48d9f40e9ccc20ea370b15)
Mark_Friedenbach comments on I played as a Gatekeeper and came pretty close to losing in a couple of occasions. Logs and a brief recap inside. - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
= f037147d6e6c911a85753b9abdedda8d){kind=logo}
Comments (46)
Missing the point. What the hell were you doing gate keeping an AI when you think AIs are universally evil?
Even the real person in this situation can lie, can't he?
The AI could simply point out that 0 and 1 are not probabilities, and now by lying you've given the AI the intellectual high ground.
Yes, but the gatekeeper may be acting several levels deep in a roleplay (roleplaying a character roleplaying another character roleplaying...etc) to pass the time and avoid emitting evidence that might allow the AI to pinpoint his preferences. The currently active character may have one of a rather large number of responses to this besides actually being more mentally pliable as a result of a loss of face (or may not even view the dialogue as a loss of face.)
It amuses me that publishing this comment will make it more challenging to implement this strategy if I elect to play as Gatekeeper again at some point in the future.
Well, to nitpick I am certain that I exist (cogito) with P(1).
Well, my confidence that I exist exceeds my confidence that probability makes sense.
If the gatekeeper really believed that he would just shut off the machine.