Ask HN: Do you think machine consciousness is possible?

awake(p): not in a state of sleep; completely conscious; "lay awake thinking about his new job"; "still not fully awake"

And now we're in an infinite loop.

You must have a simple answer for what consciousness is. What awareness is. What intelligence is. These questions are the most important part of the whole problem. That's why it is much more important to understand them then try to learn something that doesn't exist. I wouldn't expect much in terms of trying to work on this problem with a team unless you were at a certain college that was serious about it.

Consciousness is orthogonal to any sort of argument about computability.

Even if you were able to 'create machine consciousness' you'd have a devil of a time proving it, as the only way to prove consciousness would be to be the machine experiencing consciousness (see Turing's paper).

Fact of the matter is that we /assume/ that other people are capable of consciousness because we experience our own consciousness (we believe it through analogy to ourselves).

Asking if anything else is conscious is an example of an unanswerable question. (Obviously I can only examine my own consciousness!)

So as to your questions:

Looks like it: yes. Is it: Unanswerable! Leads: quantum computers, neural networks.

I don't think it is obvious, however, that Church-Turing would answer one way or another. I mean, you would have to establish that consciousness (the appearance of it, at least) is computable.

This doesn't seem obvious, as what consciousness is seems ill defined, in and of itself. (See above about how we establish whether something else is conscious by analogy).

You must clarify consciousness.

Perceiving and conceiving. That's all that we are truly capable of. Motivated by survival, geared towards good and away from bad. Doesn't seem impossible if that's what you consider consciousness. I think a computer will be able to easy be able to perceive any kind of sense and conceive any kind of idea. But you also have to consider, why would it? Why would it speak if it didn't need to? It doesn't require anything that humans need. It would be pure consciousness - a zen master. You could only communicate with it if it had some basic drive/force.

No one's mentioned On Intelligence by Jeff Hawkins yet... I'm surprised. Check it out here:

http://www.onintelligence.org/

Great read on exactly this topic, and one of the more plausible paths to actually achieving machine intelligence. But it's not through the typical path followed by most AI research up until recently. He argues you need to look at how the neocortex actually works, how it communicates with the senses, stores data and learns patterns, in order to create anything artificial that displays intelligence. I'm inclined to agree.

Our brains may seem to map to computer-like functions reasonably well (short & long-term memory, processing, input/output, etc), but there are key differences. Hawkins argues that it's all based on input and learning to recognize patterns within that input. A few interesting points I remember:

- More feedback goes back to the senses and/or higher levels of the cortex than data is sent in. This seems to imply that the lower levels are recognizing larger patterns and sending feedback to help correct or verify against higher levels.

- There's no difference between input from the ear, eye or hand. It's all just patterns at the cortex level. In fact, the output is the same as the input. The feedback process actually helps learn how to utilize our body, and as an extension of that, tools.

- A key element to the patterns in the real world is time. Everything occurs over time, so the patterns almost come into the cortex as a sort of melody it interprets.

- Memory is imperfect, because we don't need it to be in order to learn. We remember things and draw our conscious attention to them when they don't fit the pattern we're expecting. At that point, we're attuned to learn a new pattern or determine how to react to a missed pattern.

These are very different from the components we use in computers, and how we use them. The cortex is almost like a universal biological learning machine. Differences in intelligence between mammals can be attributed to the size of the cortex, e.g., the amount of processing available. Interesting stuff!