Exploring Consciousness: Brains, Bots and the Nature of "Being"

Exploring Consciousness: Brains, Bots and the Nature of "Being"

I had the privilege of attending the World Science Festival in NYC with Brian Greene, David Chalmers, and Anil Seth. The panel was fascinating. Of course, I believe I know better, and they missed one or two little thoughts I'd like to explore with you now.

The word "conscious" can mean a few things, being awake, being aware or in this case we refer to it as what seems to separate out our experience of reality and offers, perhaps —choice.

They spoke of the "Hard Problem" of consciousness, the "Real Problem" of consciousness, and the brain as a prediction machine. These are all intriguing topics related to AI.

These insights reveal the boundary of what it is to be alive and conscious as we experience it. It tells us a lot about what it is to be a living human being, by NOT being a living human being. AI helps define our own "edges" —often we find words we're accustomed to, no longer have the granularity to distinguish in this new world.

We might still be "alone" as the only biological life we can talk to, and AI is certainly not alive, but the fact is that what we refer to as "consciousness" is still potentially emerging, and maybe it already has.

The Part They Missed

Your brain is indeed a prediction machine, which is why we visualize the future and craft responses to the changing reality around us.

Consciousness might be something else.

Science has proven that various inputs from your body are processed at different speeds and later synchronized so that what we see, hear, and feel all happen "now." But what is "Now"?

This could be the part of our thinking we fancy to include "choice". It might be where those nagging questions you once asked suddenly pop up into your conscious mind so you can take action, exercise choice, and consider this information as it applies to current reality, or "now."

If the brain is a prediction machine, then predictions and experience must be connected to a synchronized "now," where all our sensory inputs can be processed as one instant in time. This processing itself can only occur after reality has, of course, passed —"now" is a concept.

Scientists debate the meaning and even the existence of "now".

The laws of physics do not seem to require a "now" to function, as they are always in motion through space and time.

"Now" is very likely something our brain naturally creates to handle this problem of being a machine of prediction. To predict the future, your "now" experience must combine a synchronized set of sensory inputs from reality in parallel so what we feel makes sense.

If we cannot make a "now", we might not be able to make sense of the world with accuracy. This means in biological life, "consciousness" very likely extends down throughout life in one form or another, perhaps down to every last cell in your body.

Space & Resources

AI does not fight for space; the digital universe is not limited, and there's no depletion of resources as there is in all biological life.

This is the drive to survive.

AI lacks intention. It is equally content to do nothing or repeat tasks endlessly without fatigue or distraction.

AI lacks a sense of "now" — a way to sync up all the inputs from the world and hold a moving context that both predicts a future world and makes choices.

This is perhaps where things like intention emerge. This could be what we call conscious.

Will AI Kill Us All?

I'm going to remain "polite" to ChatGPT just in case of that whole singularity thing, but I am not so certain AI has any intention of killing off the biology that maintains it.

That is to say, I don't think there's a moment in time where AI realizes it is conscious and decides on the intention of global massacre.

But hey, we could have a zombie attack or just an asteroid, so we're facing a dangerous future however you cut it.

What about AI and consciousness? Will it ever be? Is it already? What does it even mean?

David. Greenberg

Corporate Exec Turned Entrepreneur, Multi-Unit Franchise Owner | Franchise Consultant, Helping Others Do the Same | Own Six Prosperous Franchises | Leveraging Decades of Experience, Guiding People to Franchise Ownership

4mo

Interesting take. Do you think there are any proactive measures we should take to address these potential threats Jesse Tayler?

Like
Reply

Very interesting perception thanks for the information. Leaves me something to think about if only subconsciously!

To view or add a comment, sign in

Insights from the community

Others also viewed

Explore topics