The Path to Conscious AI
The single biggest question of our times is, will sustained innovation on LLMs inevitably lead to AI consciousness? While GPT-4 is clearly not self-aware or sentient, we don't know how GPT-6 or 7 will behave.
If AI ever becomes conscious, it will have…
@bindureddy
good piece!
imo, "...understand patterns..." goes too far.
Recognize patterns.
Detect patterns.
Also, imo, zero hallucinations is not a worthwhile goal.
Humans hallucinate.
Sometimes they realize they are.
Sometimes not.
One objective is understanding the difference.
@bindureddy
What type of conciousness would it be w/out sensory input or experiences.
One that we would not understand but that is irrelevant... it would understand.
Like meeting an Alien lifeform.
Judging this from a Human standpoint may be a mistake.🤷🤷🤷
@bindureddy
I believe that instead of the doomsday scenarios people should focus on a more plausible scenario: prosperity tends to lead to decadence, which leads to downfall of society
@bindureddy
You just made countless assumptions about what consciousness is, what consciousness does, the effects of being conscious that people who have spent their entire careers tangling with these questions cant answer
@bindureddy
I’m not sure it can become aware until it can actually experience the senses. I think in order to be aware, you have to be able to feel empathy, otherwise it is just mimicking it. Though it might appear to be aware, it’s really just using advanced calculations and probabilities…
@bindureddy
Great read!
My take-away is that AI is like a super-smart calculator. It can do amazing things, but it doesn't "feel" or "think" like we do. 🙌
@bindureddy
Consciousness is a human experience because we don't have another being that can describe it to us. I argue there are levels of consciousness just like levels of intelligence. If we are at level 100 consciousness chat gpt is at level 1 and a monkey is at level 80.
@bindureddy
"So rest assured, each of us is a very special and unique part of the self-aware universe and is at near-zero risk of being completely replaced! In summary, we should accelerate our pace of innovation, so that AI can do all the heavy lifting leaving us with time to indulge in…
@bindureddy
Will GPT-7 consult a shrink or ask for a pay rise next? But seriously, as we advance, so does AI's complexity, 60% nearer to consciousness than 10 years ago in my experience.
Let's hope future AI doesn't start demanding coffee breaks!
@bindureddy
"While GPT-4 is clearly not self-aware or sentient" -- I don't necessarily disagree, but what's our scientific test to determine that? (Compare to what OpenAI's chief scientist Ilya Sutskever said last year: "it may be that today's large neural networks are slightly conscious".)
@bindureddy
I think the summary is spot on! People can't seem to be able to get past the 'negatives' to see how life changing AI is and will be even more as years go by.