Live data from Hacker News

AI vs. AGI vs. Consciousness vs. Super-Intelligence vs. Agency

secondbreakfast.co

31–40 of 129 posts

Re: AI vs. AGI vs. Consciousness vs. Super-Intelligence vs. Agency

#31

in the now infamous Lex interview, Sam Altman proposes a test for consciousness (he attributes it to Ilya Sutskever): Somehow, create an AI by training on everything we train on now, _except_ leave out any mention of consciousness, theory of mind, cognitive science etc (maybe impossible in practice but stay with me here). Then, when the model is mature (and it is not nerf'd to avoid certain subjects) you ask it somet…

That's an inaccurate test. You can't know if the answer was real or stochastic parroting.

Any attempt at consciousness requires us to define the word. And the word itself may not even represent anything real. We have a feeling for it but those feelings could be illusions and the concept itself is loaded.

For example Love is actually a loaded concept. It's chemically induced but a lot of people attribute it to something deeper and magical. They say love is more then chemical induction.

The problem here is that for love specifically we can prove it's a mechanical concept. Straight people are romantically incapable of loving members of the same sex. So the depth and the magic of it all is strictly segmented based off of biological sex? Doesn't seem deep or meaningful at all. Thus love is an illusion. A loaded and mechanic instinct tricking us with illusions of deeper meaning and emotions into creating progeny for future generations.

Consciousness could be similar. We feel there is something there, but really there isn't.

Re: AI vs. AGI vs. Consciousness vs. Super-Intelligence vs. Agency

#32
How are people defining agency? Because GPT-4 can have agency, it just needs to be put in specific situations to have that agency.

For example, I could theoretically hook up my Home Assistant instance to GPT-4 and run a script every 10 minutes telling GPT-4 the temperature and asking for a yes or no response to whether I should turn on the AC or heat. That sounds to me like the AI now has agency over the temperature in my home. You don't even need any real AI for this. Google's Nests have some algorithm that adjust temperature based off usage.

Is this not agency? Or is the author not counting agency without consciousness as agency?

Re: AI vs. AGI vs. Consciousness vs. Super-Intelligence vs. Agency

#33

in the now infamous Lex interview, Sam Altman proposes a test for consciousness (he attributes it to Ilya Sutskever): Somehow, create an AI by training on everything we train on now, _except_ leave out any mention of consciousness, theory of mind, cognitive science etc (maybe impossible in practice but stay with me here). Then, when the model is mature (and it is not nerf'd to avoid certain subjects) you ask it somet…

That brings up a lot of hard questions. Supposing you had that AI but didn't allow it to churn in the background when not working on a problem. Human brains don't stop. They constantly process data in both conscious and unconscious ways. The AIs we've built don't do that. The meaning of the concept of "self" for a human is something a huge percentage of their thoughts interact with directly or indirectly. Will an AI ever develop a similar concept if it never has to chew on the problem for a long period?

Re: AI vs. AGI vs. Consciousness vs. Super-Intelligence vs. Agency

#34
post #32

How are people defining agency? Because GPT-4 can have agency, it just needs to be put in specific situations to have that agency. For example, I could theoretically hook up my Home Assistant instance to GPT-4 and run a script every 10 minutes telling GPT-4 the temperature and asking for a yes or no response to whether I should turn on the AC or heat. That sounds to me like the AI now has agency over the temperature…

By this logic any random (or maybe even non-random, but I suspect that that case is a bit more boring) process has agency.

Re: AI vs. AGI vs. Consciousness vs. Super-Intelligence vs. Agency

#35

> Almost all AGI doomsayers assume AGI will have agency. They have this vision of the machine deciding it’s time to end civilization. No. Agency is not a necessary condition for AI to do massive damage. I don't believe agency is really well-defined either. An AI merely needs to be hooked up to enough physical systems, have sufficiently complex reaction mechanisms, and some way of looping to do a lot of damage. For th…

You can imbue an LLM with agency. Actual "do whatever you want" agency. It's not hard. In fact it's already being done in a few projects. You can embody an LLM too. This too is not hard. Cost is probably the most prohibitive thing.

This is all nonsense. You could argue that it's possible (which certainly has not been shown to be true) but it certainly not true to say it is "not hard".

If there are LLM projects with "do whatever you want" agency - please provide links.

Re: AI vs. AGI vs. Consciousness vs. Super-Intelligence vs. Agency

#36
post #34
post #32

How are people defining agency? Because GPT-4 can have agency, it just needs to be put in specific situations to have that agency. For example, I could theoretically hook up my Home Assistant instance to GPT-4 and run a script every 10 minutes telling GPT-4 the temperature and asking for a yes or no response to whether I should turn on the AC or heat. That sounds to me like the AI now has agency over the temperature…

By this logic any random (or maybe even non-random, but I suspect that that case is a bit more boring) process has agency.

Yes, which is why I think agency is a bad metric for judging the progress of AI.

Re: AI vs. AGI vs. Consciousness vs. Super-Intelligence vs. Agency

#37
post #32

How are people defining agency? Because GPT-4 can have agency, it just needs to be put in specific situations to have that agency. For example, I could theoretically hook up my Home Assistant instance to GPT-4 and run a script every 10 minutes telling GPT-4 the temperature and asking for a yes or no response to whether I should turn on the AC or heat. That sounds to me like the AI now has agency over the temperature…

Wouldn't agency involve ChatGPT deciding it wanted to do this, instead of you telling it to do it?

Re: AI vs. AGI vs. Consciousness vs. Super-Intelligence vs. Agency

#38

in the now infamous Lex interview, Sam Altman proposes a test for consciousness (he attributes it to Ilya Sutskever): Somehow, create an AI by training on everything we train on now, _except_ leave out any mention of consciousness, theory of mind, cognitive science etc (maybe impossible in practice but stay with me here). Then, when the model is mature (and it is not nerf'd to avoid certain subjects) you ask it somet…

That brings up a lot of hard questions. Supposing you had that AI but didn't allow it to churn in the background when not working on a problem. Human brains don't stop. They constantly process data in both conscious and unconscious ways. The AIs we've built don't do that. The meaning of the concept of "self" for a human is something a huge percentage of their thoughts interact with directly or indirectly. Will an AI…

Haven't thought of this. Couldn't you just give a model a way to constantly "chew" on something? Maybe an ever-ending loop of some sort of prompt stimulation?

Re: AI vs. AGI vs. Consciousness vs. Super-Intelligence vs. Agency

#39

in the now infamous Lex interview, Sam Altman proposes a test for consciousness (he attributes it to Ilya Sutskever): Somehow, create an AI by training on everything we train on now, _except_ leave out any mention of consciousness, theory of mind, cognitive science etc (maybe impossible in practice but stay with me here). Then, when the model is mature (and it is not nerf'd to avoid certain subjects) you ask it somet…

Flicks the off switch

Re: AI vs. AGI vs. Consciousness vs. Super-Intelligence vs. Agency

#40
post #32

How are people defining agency? Because GPT-4 can have agency, it just needs to be put in specific situations to have that agency. For example, I could theoretically hook up my Home Assistant instance to GPT-4 and run a script every 10 minutes telling GPT-4 the temperature and asking for a yes or no response to whether I should turn on the AC or heat. That sounds to me like the AI now has agency over the temperature…

[Author here] I really have no definition of agency. Do you know of any good ones?
Post reply on HN