A putative new idea for AI control; index here.

In previous posts, I posited AIs caring only about virtual worlds - in fact, being defined as processes in virtual worlds, similarly to cousin_it's idea. How could this go? We would want the AI to reject offers of outside help - be they ways of modifying its virtual world, or ways of giving it extra resources.

Let V be a virtual world, over which a utility function u is defined. The world accepts a single input string O. Let P be a complete specification of an algorithm, including the virtual machine it is run on, the amount of memory it has access to, and so on.

Fix some threshold T for u (to avoid the the subtle weeds of maximising). Define the statement:

r(P,O,V,T): "P(V) returns O, and either E(u|O)>T or O=∅"

And the string valued program:

Q(V,P,T): "If you can find that there exists a non-empty O such that r(P,O,V,T), return O. Else return ∅."

Here "find" and "E" are where the magic-super-intelligence-stuff happens.

Now, it seems to me that Q(V,Q,T) is the program we are looking for. It is uninterested in offers to modify the virtual world, because E(u|O)>T is defined over the unmodified virtual world. We can set it up so that the first thing it proves is something like "If I (ie Q) prove E(u|O)>T, then r(Q,O,V,T)." If we offer it more computing resources, it can no longer make use of that assumption, because "I" will no longer be Q.

Does this seem like a possible way of phrasing the self-containing requirements? For the moment, this seems to make it reject small offers of extra resources, and be indifferent to large offers.

A putative new idea for AI control; index here.

In a previous post, I talked about an AI operating only on a virtual world (ideas like this used to be popular, until it was realised the AI might still want to take control of the real world to affect the virtual world; however, with methods like indifference, we can guard against this much better).

I mentioned that the more of the AI's algorithm that existed in the virtual world, the better it was. But why not go the whole way? Some people at MIRI and other places are working on agents modelling themselves within the real world. Why not have the AI model itself as an agent inside the virtual world? We can quine to do this, for example.

Then all the restrictions on the AI - memory capacity, speed, available options - can be specified precisely, within the algorithm itself. It will only have the resources of the virtual world to achieve its goals, and this will be specified within it. We could define a "break" in the virtual world (ie any outside interference that the AI could cause, were it to hack us to affect its virtual world) as something that would penalise the AI's achievements, or simply as something impossible according to its model or beliefs. It would really be a case of "given these clear restrictions, find the best approach you can to achieve these goals in this specific world".

It would be idea if the AI's motives were not given in terms of achieving anything in the virtual world, but in terms of making the decisions that, subject to the given restrictions, were most likely to achieve something if the virtual world were run in its entirety. That way the AI wouldn't care if the virtual world were shut down or anything similar. It should only seek to self modify in way that makes sense within the world, and understand itself existing completely within these limitations.

Of course, this would ideally require flawless implementation of the code; we don't want bugs developing in the virtual world that point to real world effects (unless we're really confident we have properly coded the "care only about the what would happen in the virtual world, not what actually does happen).

Any thoughts on this idea?

Do Virtual Humans deserve human rights?

Slate Article

I think the idea of storing our minds in a machine so that we can keep on "living" (and I use that term loosely) is fascinating and certainly and oft discussed topic around here. However, in thinking about keeping our brains on a hard drive we have to think about rights and how that all works together. Indeed the technology may be here before we know it so I think its important to think about mindclones. If I create a little version of myself that can answer my emails for me, can I delete him when I'm done with him or just turn him in for a new model like I do iPhones? 

I look forward to the discussion.

A new study indicates that people become more utilitarian (save more lives) when viewing a moral dilemma in a virtual reality situation, as compared to reading the same situation in text.

Abstract.

Although research in moral psychology in the last decade has relied heavily on hypothetical moral dilemmas and has been effective in understanding moral judgment, how these judgments translate into behaviors remains a largely unexplored issue due to the harmful nature of the acts involved. To study this link, we follow a new approach based on a desktop virtual reality environment. In our within-subjects experiment, participants exhibited an order-dependent judgment-behavior discrepancy across temporally-separated sessions, with many of them behaving in utilitarian manner in virtual reality dilemmas despite their non-utilitarian judgments for the same dilemmas in textual descriptions. This change in decisions reflected in the autonomic arousal of participants, with dilemmas in virtual reality being perceived more emotionally arousing than the ones in text, after controlling for general differences between the two presentation modalities (virtual reality vs. text). This suggests that moral decision-making in hypothetical moral dilemmas is susceptible to contextual saliency of the presentation of these dilemmas.

Full paper

Video of simulations

I would like to consolidate LW members in a new way. I believe we can organize an on-line gathering in some form, for example as a set of chat rooms to discuss different topics in the real time. This event can be announced in advance to help everyone to arrange plans. I hope the discussion can be more intense and productive than in chats that open for prolonged periods of time. And comparing to the lesswrong.com this event should give some feeling of a real conversation, which I can not get while posting articles and comments here.

If you have any additions, ideas and proposals please let me know.

Doing some insomniac reading of the Quantum Sequence, I think that I've gotten a reasonable grasp of the principles of decoherence, non-interacting bundles of amplitude, etc. I then tried to put that knowledge to work by comparing it with my understanding of virtual particles (whose rate of creation in any area is essentially equivalent to the electromagnetic field), and I had a thought I can't seem to find mentioned elsewhere.

If I understand decoherence right, then quantum events which can't be differentiated from each other get summed together into the same blob of amplitude. Most virtual particles which appear and rapidly disappear do so in ways that can't be detected, let alone distinguished. This seems as if it could potentially imply that the extreme evenness of a vacuum might have to do more with the overall blob of amplitude of the vacuum being smeared out among all the equally-likely vacuum fluctuations, than it does directly with the evenness of the rate of vacuum fluctuations themselves. It also seems possible that there could be some clever way to test for an overall background smear of amplitude, though I'm not awake enough to figure one out just now. (My imagination has thrown out the phrase 'collapse of the vacuum state', but I'm betting that that's just unrelated quantum buzzword bingo.)

Does anything similar to what I've just described have any correlation with actual quantum theory, or will I awaken to discover all my points have been voted away due to this being complete and utter nonsense?