RE: [agi] Why is multiple superintelligent AGI's safer than a single AGI?
Ben, In reply to my para saying : if the one AGI goes feral the rest of us are going to need to access the power of some pretty powerful AGIs to contain/manage the feral one. Humans have the advantage of numbers but in the end we may not have the intellectual power or speed to counter an AGI that is actively setting out to threaten humans. you said: I don't see why multiple superintelligent AGI's are safer than a single one [EMAIL PROTECTED]@[EMAIL PROTECTED] I thought the previous para gave at least one good reason. :) That paragraph gave one possible dynamic in a society of AGI's, but there are many many other possible social dynamics, including those that lead to mobs of rampaging violent AGI's. What's the probability distribution across the different AGI social dynamic patterns??? Certainly among human societies, the only analogue we have, Human societies are NOT the only analogues we have so for understanding GI social behaviour. There are lots of social animals and chimpanzees and bonobos are very closely related to humans (they are moreclosely realted to humans than they are to gorillas, orangutans and gibbons). Well, OK, but I reckon ape societies are worse analogues for AGI societies than human societies are... So I think the way to approach the issue is to say what behavioural drivers do we need to generate in AGIs so that their collective behaviour tend strongly towards peaceful or better still peace making behaviour. Sure... I'm just not so sure that there's any benefit to the society of AGI as opposed to one big AGI approach ben --- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
RE: [agi] Why is multiple superintelligent AGI's safer than a single AGI?
Ben, Ben: That paragraph gave one possible dynamic in a society of AGI's, but there are many many other possible social dynamics Of course. What you say is quite true. But so what? Let's go back to that one possible dynamic. Can't you bring yourself to agree that if a one-and-only super-AGI went feral that humans would then be at a greater disadvantage relative to it than if there was more than one AGI around and the humans could call on the help of one or more of the other AGIs?? Forget about all the other possible hypotheticals. Is my assessment of the specific scenario above about right or not - doesn't it have some element of common sense about it? If there is any benefit in having more than one AGI around in the case where an AGI does go feral then your comment I'm just not so sure that there's any benefit to the society of AGI as opposed to one big AGI approach no longer holds as an absolute. It then gets back to having a society of AGIs might be an advantage in certain cercumstances, but having more than one AGI might have the following down sides. At this point a balanced risk/benefit assessment can be made (not definitive of course since we haven't seen super-intelligent AGIs operation yet). But at least we've got some relevant issues on the table to think about. Cheers, Philip
Re: [agi] Why is multiple superintelligent AGI's safer than a single AGI?
Hi Eliezer, This does not follow. If an AI has a P chance of going feral, then a society of AIs may have P chance of all simultaneously going feral I can see you point but I don't agree with it. If General Motors churns out 100,000 identical cars with all the same charcteristics and potiential flaws, they will not all fail at the same instant in time. Each of them will be placed in a different operating environment and the failures will probably spread over a bell curve style distribution. If we apply this logic to AGIs we have a chance to enlist the support of most of the AGIs to 'recall' the population to take preventive action to avoid failure and will have their help to deal with the AGIs that have already failed. Cheers, Philip
RE: [agi] Why is multiple superintelligent AGI's safer than a single AGI?
Ben Goertzel wrote: Yes, I see your point now. If an AI has a percentage p chance of going feral, then in the case of a society of AI's, only p percent of them will go feral, and the odds are that other AI's will be able to stop it from doing anything bad. But in the case of only one AI, then there's just a p% chance of it going feral, without much to do about it... Unknowns are the odds of AI's going feral and supersmart at the same time, and the effects of society-size on the probability of ferality... But you do have a reasonable point, I'll admit, and I'll think about it more... This does not follow. If an AI has a P chance of going feral, then a society of AIs may have P chance of all simultaneously going feral - it depends on how much of the probability is independent among different AIs. yeah, this dependency is what I meant by my awkward and hasty phrase the effects of society-size on the probability of going feral. (I should have said sociodynamics, not society-size). This is why Philip's point is reasonable but not well-demonstrated... Actually, for the most worrisome factors, such as theoretical flaws in the theory of AI morality, I would expect the risk factor to be almost completely shared among all AIs. Furthermore, risk factors stemming from divergent rates of self-enhancement or critical thresholds in self-enhancement may not be at all improved by multiply copying or multiply diverging AIs if AI improvement is less than perfectly synchronized. All true. There are a lot of uncertainties here. My guess is that I'm gonna want to build one big Novamente mind and not a society of smaller ones. But I'll revisit Philip's points very carefully in a few years when the time comes to think about such things seriously... Ben G --- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
RE: [agi] Why is multiple superintelligent AGI's safer than a single AGI?
Eliezer is certainly correct here -- your analogy ignores probabilistic dependency, which is crucial. Ben -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Eliezer S. Yudkowsky Sent: Tuesday, March 04, 2003 1:33 AM To: [EMAIL PROTECTED] Subject: Re: [agi] Why is multiple superintelligent AGI's safer than a single AGI? Philip Sutton wrote: Hi Eliezer, This does not follow. If an AI has a P chance of going feral, then a society of AIs may have P chance of all simultaneously going feral I can see you point but I don't agree with it. If General Motors churns out 100,000 identical cars with all the same charcteristics and potiential flaws, they will */not /*all fail at the same instant in time. Each of them will be placed in a different operating environment and the failures will probably spread over a bell curve style distribution. That's because your view of this problem has automatically factored out all the common variables. All GM cars fail when dropped off a cliff. All GM cars fail when crashed at 120 mph. All GM cars fail on the moon, in space, underwater, in a five-dimensional universe. All GM cars are, under certain circumstances, inferior to telecommuting. How much of the risk factor in AI morality is concentrated into such universals? As far as I can tell, practically all of it. Every AI morality failure I have ever spotted has been of a kind where a society of such AIs would fail in the same way. The bell-curve failures to which you refer stem from GM making a cost-performance tradeoff. The bell-curve distributed failures, like the fuel filter being clogged or whatever, are *acceptable* failures, not existential risks. It therefore makes sense to accept a probability X of failure, for component Q, which can be repaired at cost C when it fails; and when you add up all those probability factors you end up with a bell curve. But if the car absolutely had to work, you would be minimizing X like hell, to the greatest degree allowed by your *design ability and imagination*. You'd use a diamondoid fuel filter. You'd use three of them. You wouldn't design a car that had a single point of failure at the fuel filter. You would start seriously questioning whether what you really wanted should be described as a car. Which in turn would shift the most probable cause of catastrophic failure away from bell-curve probabilistic failures and into outside-context failures of imagination. -- Eliezer S. Yudkowsky http://singinst.org/ Research Fellow, Singularity Institute for Artificial Intelligence --- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED] --- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
Re: [agi] Why is multiple superintelligent AGI's safer than a single AGI?
Eliezer, That's because your view of this problem has automatically factored out all the common variables. All GM cars fail when dropped off a cliff. All GM cars fail when crashed at 120 mph. All GM cars fail on the moon, in space, underwater, in a five-dimensional universe. All GM cars are, under certain circumstances, inferior to telecommuting. Good point. Although not all failires will be of this sort so the group strategy is still useful for at least a sibset of the failure cases. Seems to me then that safety lies in a combination of all our best safety factors: - designing all AGIs to be as effectively friendly as possible - as if we had a one shot chance of getting it right and we can't afford the risk of failure, and AS WELL - developing quite a few different types of AGI architecture so that the risk of sharing the same class of critical error is reduced; and AS WELL - having a society of AGIs with multiples of each different type - that are uniquely trained - so that the degree of sameness and hence risk of failure is not so tightly linked. Cheers, Philip --- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED]