On Monday 01 October 2007 11:34:09 am, Richard Loosemore wrote: > Right, now consider the nature of the design I propose: the > motivational system never has an opportunity for a point failure: > everything that happens is multiply-constrained (and on a massive scale: > far more than is the case even in our own brains). Once the system is > set up to behave according to a diffuse set of checks and balances (tens > of thousands of ideas about what is "right", rather than one single > directive), it can never wander far from that set of constraints without > noticing the departure immediately.
That's essentially what I've been proposing, although the form appears different. Namely, design AIs so that they form a society, with the same kind of ability to monitor and police each other that human societies give us. This has the advantage that if enough of us do it with our AIs, it won't matter (at least it won't be catastrophic) that some other people create psychopathic ones. It's clearly valuable to think about how to do this inside a single system, but crucial to make sure we can do it for the population of all AIs as a whole. Josh ----- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=8660244&id_secret=48561613-4de0f2
