[yannyu]

If AI has a survival instinct, then we should theoretically see evidence of it if we construct the right environment for AI to express it. Animals and cellular organisms demonstrate a survival instinct under the right conditions, so we would have to find equivalent conditions for a hypothetical machine intelligence.

Conversely, we know that if we take animals that do have a survival instinct and put them into the wrong kinds of environments, they will not thrive and will degenerate or possibly commit suicide. Similarly, if AI did have a survival instinct, do we think we've created an environment where that could be reasonably tested and observed?

[drxzcl]

I can make an AI system with a survival instinct right now. Of course, all that will do is make people tell me “it’s not a proper survival instinct” or move the goal posts and tell me I need yet some other property.

This whole endeavor is doomed from the beginning. There is no crucial test for “consciousness”, just ad hoc criteria people come up with to land on the conclusions that leave their belief system intact.

Consciousness is not a concept that can be rendered operational.

[Ekaros]

I can make state machine that acts like it has survival instinct. But it certainly isn't something we would consider conscious. So I am not exactly sure how good most tests are.

[drxzcl]

But what would we consider conscious?

My position is that there is no actual, definitive answer to that question, and therefore it makes no sense engaging with the concept.

[FrustratedMonky]

That is entire plot of 'Ex Machina'.

There are plenty of people that say AI has already displayed a survival instinct, by threatening users if they talk about shutting it down. Or to use a market or blackmail, to get funds to source an external machine to run on.

There are bunch of articles proclaiming AI is trying to break out. Can't find a real study on it.

https://www.wsj.com/opinion/ai-is-learning-to-escape-human-c...