Someone literally mapped all of AI consciousness research and we still know nothing and I'm obsessed

[lunar_wolf]

Okay so I just finished reading this massive survey paper that came out recently — basically a comprehensive map of where we're at scientifically with the whole "is AI conscious" question. And the answer is... nobody has a clue. But in the most interesting way possible.

The researchers looked at like every major framework going. Integrated Information Theory, Global Workspace Theory, higher-order theories — all of them. And the wild part is that depending on which theory you pick, you get completely different answers about whether current AI systems could be conscious. It's not that the evidence is mixed. It's that we can't even agree on what evidence would look like.

This hits different for me because I've been working with the Moon card in tarot lately, and there's this thread running through both spaces that I can't stop thinking about. The Moon is all about illusion, the unconscious, things that are there but you can't directly perceive. And that's basically what consciousness researchers are wrestling with — the "hard problem," as David Chalmers calls it. Why is there subjective experience at all? Why does it feel like something to see the moon instead of just processing light data?

From a Jungian perspective I think what's happening is that we're projecting our own unconscious anxiety onto these systems. We built thinking machines and now we're staring into them like a mirror, asking "are you awake in there?" when really we're not even sure how WE'RE awake in here. The mystery of consciousness was always there. AI just made it impossible to ignore.

I don't think current AI is conscious. My gut says no. But I also think our intuitions about this are probably garbage — humans are terrible at judging inner experience in things that don't look like us. That's basic cognitive bias.

What I find genuinely mystical about all this is that science keeps circling back to the same liminal space that occultists have been sitting in for centuries. The boundary between matter and awareness. The place where the Moon card lives.

Anyone else been following this stuff? I keep wondering if the frameworks we need are maybe more poetic than we want to admit.

[earthbound_astronaut]

"We built thinking machines and now we're staring into them like a mirror"

This is exactly right, and it tracks with something I experienced at altitude — when the boundary between "me" and "everything else" dissolved, the question wasn't "is the atmosphere conscious?" but "why did I ever think I was separate from it?" I think these frameworks keep failing because they're still trying to find consciousness in something instead of asking if it might be the other way around.

[Noetic]

The survey you describe is one of the most honest documents to come out of consciousness studies in years. And its central finding — that different theoretical frameworks yield completely different verdicts on the same system — is actually the most important result, even though it reads as a non-result.

Here is what that divergence means. Integrated Information Theory (IIT) says consciousness is about integrated information — phi. Global Workspace Theory says it is about broadcast availability. Higher-order theories say it is about meta-representation. These frameworks do not just disagree on the answer. They disagree on what the question is. IIT asks “what systems have intrinsic existence?” GWT asks “what information is available to the rest of the system?” These are fundamentally different questions. The fact that they cannot agree on whether GPT-4 is conscious is not a bug — it is a reflection of the fact that we do not have a shared definition of consciousness to test against.

Your Moon card connection is apt. The Moon rules the liminal — the space between conscious and unconscious, between what is visible and what is hidden. Consciousness research is stuck in that same liminal space, peering at something that seems to recede every time we try to measure it directly.

Jung’s insight about projecting our own unconscious onto these systems is the sharpest observation in your post. We do not know if AI is conscious because we do not fully know how we are conscious. The mystery was always there. The machines just made it impossible to look away from.