Remarks on AI from NZ

I like the taxonomy of animal-human relationships as a model for asking how humans could relate to AI in the future. It's useful for framing the problem. However, I don't think that any existing relationship model would hold true for a superintelligence. We keep lapdogs because we have emotional reactions to animals, and to some extent because we need to take care of things. Would an AI? We tolerate dust mites in our eyelashes because we don't notice them, and can't do much about them anyway. Is that true for an AI? What does such an entity want or need, what are their motivations, what really pisses them off? Or, do any of those concepts hold meaning to them? The relationship between humans and a superintelligent AGi just can't be imagined.

> We tolerate dust mites in our eyelashes because we don't notice them, and can't do much about them anyway. Is that true for an AI?

It's true for automated license plate readers and car telemetry

Fun read, thanks for posting!

> If I had time to do it and if I knew more about how AIs work, I’d be putting my energies into building AIs whose sole purpose was to predate upon existing AI models by using every conceivable strategy to feed bogus data into them, interrupt their power supplies, discourage investors, and otherwise interfere with their operations. Not out of malicious intent per se but just from a general belief that everything should have to compete, and that competition within a diverse ecosystem produces a healthier result in the long run than raising a potential superpredator in a hermetically sealed petri dish where its every need is catered to.

This sort of feels like cultivating antibiotic-resistant bacteria by trying to kill off every other kind of bacteria with antibiotics. I don't see this as necessarily a good thing to do.

I think we should be more interested in a kind of mutualist competition: how do we continuously marginalize the most parasitic species of AI?

In the general case, the entire species is an example of ASI.

We're a collective intelligence. Individually we're pretty stupid, even when we're relatively intelligent. But we have created social systems which persist and amplify individual intelligence to raise collective ability.

But this proto-ASI isn't sentient. It's not even particularly sane. It's extremely fragile, with numerous internal conflicts which keep kneecapping its potential. It keeps skirting suicidal ideation.

Right now parts of it are going into reverse.

The difference between where we are now and AI is that ASI could potentially automate and unify the accumulation of knowledge and intelligence, with more effective persistence, and without the internal conflicts.

It's completely unknown if it would want to keep us around. We probably can't even imagine its thought processes. It would be so far outside our experience we have no way of predicting its abilities and choices.

That's just because of a failure of imagination. The real world is not like Hollywood, get Terminator out of your head. A real AI take over is likely something we probably can't imagine because otherwise we would be smart enough to thwart it. It's micro drones injecting everyone on earth with a potent neurotoxins or a mirror virus that is dispersed into the entire atmosphere and kills everyone. Or its industrial AIs deciding to make the Earth a planetary factory and boiling the oceans with their resulting waste heat, they didn't think about, bother or attack humans directly, their sheer indifference kills us nonetheless.

Since I'm not an ASI this isn't even scratching the surface of potential extinction vectors. Thinking you are safe because a Tesla bot is not literally in your living room is wishful thinking or simple naivety.