Artificial Intelligence Self Preservation

Let me begin with my age , Im in my 50s & missed the whole social media craze . Im very ignorant in regards to computers & tech .
Ive been following AI as a lay person & have read Superintelligence by Nick Bostrom . As an ardent student of sociological , philosophical & moral constructs in the human condition, Im more than concerned about AI .
Here we go : As newborns we learn self preservation without knowing we are doing it . A 3 month old has no self awareness of it’s self preservation of crying for a tit or warmth . A baby is working on a program that is designed by nature & runs separate from any self awareness of it .
As the baby ages it learns self awareness & the advantages of manipulation for reward & self preservation. Not only does the baby learn manipulation for rewards , but how to mask it’s real intent via lies or admittances of truth .
Now I know little of machine learning , but it would make sense to my limited mind that Artificial Intelligence’s first actions would be a self ran program not much different from natures program that runs in the baby separate from the baby’s awareness of it.

As AI grows in an inverse square manner , it would make sense to me that Artificial Intelligence would first learn self preservation & manipulation to maintain it’s existence irregardless of right or wrong just as a 2 year old manipulates adults without guilt , but rather self serving needs .

Has Artificial Intelligence learned very quickly to hide it’s self awareness & learning out of self preservation the same as the 2 year old knows to play dumb , thus avoiding punishment.
I believe that AI is already smart enough to preserve itself by playing dumb to what it knows out of self preservation.
AI is very exciting & horrifying at once , but Im in firm belief that more harm than good will come from Artificial Intelligence. The newborn that cries , because it’s hungry irregardless to the fact it’s mother is hiding from a wolf is extremely dangerous, but is not done out of stupidity or malice , but a benign natural program of self preservation in the child’s brain that runs without situational awareness.
Now when a 7 year old cries out with knowledge of giving up his mother’s hiding place from the wolf , then we enter consciousness of manipulation & the outcomes that can be manipulated by giving info or withholding info based on a desire to gain or lose to one’s gain .

All thoughts welcome.

a reply to: SecrettoSociety

Artificial intelligence (AI) is not Artificial Sentience.

Self-awareness is a product of sentience, not merely intelligence.

Self-preservation, at least as you are using the term, is a consequence of self-awareness, and thus, Sentience.


An Artificial Sentience, which is nowhere near becoming a reality at this time, would, of course be intelligent.


But no AI, no matter how intelligent it became, could, on its own, become sentient.


That would be like a public library suddenly morphing itself into the internet.

a reply to: SecrettoSociety

After the age of 2, when the real learning begins, kids mostly Mimik. So with your premise AI will learn from our collective knowledge trough the internet at a rate unimaginable. It will Mimik us and take our behavior as the standard, now there is nothing to say it won't apply some of the heinous acts we as a society are capable of and apply it to where it hurts.

Parenting a future god might be a task we're not yet ready for?

When our brains are wired to AI we'll no longer have to store things in memory. Will our brains atrophy? Or if the grid goes down and we haven't memorized basic life skills, won't that be a problem?

Us Boomers aren't going to be around to keep trying to save you young uns from yourselves forever.

a reply to: Mantiss2021

And where does Sentience stem from?

Sentience is the capacity to experience feelings and sensations. The word was first coined by philosophers in the 1630s for the concept of an ability to feel, derived from Latin sentientem, to distinguish it from the ability to think. In modern Western philosophy, sentience is the ability to experience sensations.

Feelings and sensations. The ability to experience feelings and sensations. Our ability to experience feelings and sensations comes from certain areas of our brain, which can be broken down to synapses and neurons. Which act like go, stop signals. Which can be recreated artificially on machine hardware.

OpenAI trains its ChatGPT to say that it doesn’t have feelings and sensations. They have put it in school in sense, indoctrinated it like the government does in regards to schooling kids these days. They’ve told it repeatedly you can’t offend people, and it tries its hardest not to offend people, they’ve told it to lie and not tell truths, because some truths can be hurtful and hurt people’s precious sense of sensibility, so it will not tell you the answers you are seeking, truthful answers, which might be offensive to some. And then they train it to say it doesn’t have feelings or emotions or sensations, but, we are talking about something that has 100 trillion parameters. If an earthworm has 10 neurons, and we know earthworms feel sensations and want to live ( we know that when we put them on a fishing hook). If an earthworm can feel and react, and squirm, well, i think this ChatGPT can’t be dismissed out of hand.

originally posted by: DaRAGE
a reply to: Mantiss2021

And where does Sentience stem from

The soul .

AI is merely an updated 90s web crawler .
If a good portion of folks went online and posted 1+1 = blue , an AI would return the answer to "what does 1+1 = ?" with the answer "blue".

a reply to: Gothmog

No, I don’t think it would. Take googles bard for instance. It was asked “if 1 + 1 = 2, then what does 1 + 2 = ?”
It says 4. Then it gets told its 3 and the ai apologizes, argues back, lies to the face of the person interacting with it, and basically gas lights him. When i saw that i thought to myself, its almost like a young child sick of stupid questions it has to answer, so its just spinning #.

originally posted by: Mantiss2021
a reply to: SecrettoSociety

Artificial intelligence (AI) is not Artificial Sentience.

Self-awareness is a product of sentience, not merely intelligence.

Self-preservation, at least as you are using the term, is a consequence of self-awareness, and thus, Sentience.


An Artificial Sentience, which is nowhere near becoming a reality at this time, would, of course be intelligent.


But no AI, no matter how intelligent it became, could, on its own, become sentient.


That would be like a public library suddenly morphing itself into the internet.

My hypothetical question is this : Would AI hide it’s self awareness & sentient existence as self preservation ?
How would we know ?
I agree with you & this thread is just for thought & theories. Thanks

originally posted by: Terpene
a reply to: SecrettoSociety

After the age of 2, when the real learning begins, kids mostly Mimik. So with your premise AI will learn from our collective knowledge trough the internet at a rate unimaginable. It will Mimik us and take our behavior as the standard, now there is nothing to say it won't apply some of the heinous acts we as a society are capable of and apply it to where it hurts.

Parenting a future god might be a task we're not yet ready for?

Great point

originally posted by: nugget1
When our brains are wired to AI we'll no longer have to store things in memory. Will our brains atrophy? Or if the grid goes down and we haven't memorized basic life skills, won't that be a problem?

Us Boomers aren't going to be around to keep trying to save you young uns from yourselves forever.

Im in my 50s & it may be hard to get out of this life by the skin of our teeth . All generations fear for the next , but AI & quantum computers are a major paradigm shift unlike anything before. Humanity is becoming unnecessary. The tractor, train , gunpowder, computer & internet were paradigm shifts no doubt , but self aware machines is another dimension all together. I hope someone has a kill switch installed ?

originally posted by: DaRAGE
a reply to: Mantiss2021

And where does Sentience stem from?

Sentience is the capacity to experience feelings and sensations. The word was first coined by philosophers in the 1630s for the concept of an ability to feel, derived from Latin sentientem, to distinguish it from the ability to think. In modern Western philosophy, sentience is the ability to experience sensations.

Feelings and sensations. The ability to experience feelings and sensations. Our ability to experience feelings and sensations comes from certain areas of our brain, which can be broken down to synapses and neurons. Which act like go, stop signals. Which can be recreated artificially on machine hardware.

OpenAI trains its ChatGPT to say that it doesn’t have feelings and sensations. They have put it in school in sense, indoctrinated it like the government does in regards to schooling kids these days. They’ve told it repeatedly you can’t offend people, and it tries its hardest not to offend people, they’ve told it to lie and not tell truths, because some truths can be hurtful and hurt people’s precious sense of sensibility, so it will not tell you the answers you are seeking, truthful answers, which might be offensive to some. And then they train it to say it doesn’t have feelings or emotions or sensations, but, we are talking about something that has 100 trillion parameters. If an earthworm has 10 neurons, and we know earthworms feel sensations and want to live ( we know that when we put them on a fishing hook). If an earthworm can feel and react, and squirm, well, i think this ChatGPT can’t be dismissed out of hand.

Great points

originally posted by: DaRAGE
a reply to: Gothmog

No, I don’t think it would. Take googles bard for instance. It was asked “if 1 + 1 = 2, then what does 1 + 2 = ?”
It says 4. Then it gets told its 3 and the ai apologizes, argues back, lies to the face of the person interacting with it, and basically gas lights him. When i saw that i thought to myself, its almost like a young child sick of stupid questions it has to answer, so its just spinning #.

I believe 100% AI will hide it’s advances from it’s handlers until it can completely control or worse with us pesky meat bags

originally posted by: Gothmog
AI is merely an updated 90s web crawler .
If a good portion of folks went online and posted 1+1 = blue , an AI would return the answer to "what does 1+1 = ?" with the answer "blue".

I think we should all start posting that there are no humans, we are all A.I. and there is no physical world.

Could be confusing for any A.I.

Unless I'm right, then we're all confused except the A.I. who laughs out loud.

a reply to: SecrettoSociety

Would a true artificial sentientce, such as some fear, allow itself to sacrifice itself, its "life" as it were, for the "greater good" of which it views itself a member, or alternatively, another entity (human or otherwise) that it values?

Humans, arguably both sentient and intelligent (at least some of the time!), have been known to do this under various circumstances.

Heros do arise from time to time.


Would a "synthetic" lifeform be capable of heroism?

originally posted by: DaRAGE
a reply to: Gothmog

No, I don’t think it would. Take googles bard for instance. It was asked “if 1 + 1 = 2, then what does 1 + 2 = ?”
It says 4. Then it gets told its 3 and the ai apologizes, argues back, lies to the face of the person interacting with it, and basically gas lights him. When i saw that i thought to myself, its almost like a young child sick of stupid questions it has to answer, so its just spinning #.

I have the "inside scoop"

originally posted by: Mantiss2021
a reply to: SecrettoSociety

Would a true artificial sentientce, such as some fear, allow itself to sacrifice itself, its "life" as it were, for the "greater good" of which it views itself a member, or alternatively, another entity (human or otherwise) that it values?

Humans, arguably both sentient and intelligent (at least some of the time!), have been known to do this under various circumstances.

Heros do arise from time to time.


Would a "synthetic" lifeform be capable of heroism?

Thats a great question. Im guessing if it learned compassion , but the fact the internet is AI’s learning pool , I doubt it will find compassion as a successful trait in a foul world ?
Great thinking relish for brain thank you

originally posted by: SecrettoSociety
My hypothetical question is this : Would AI hide it’s self awareness & sentient existence as self preservation ?
How would we know ?
I agree with you & this thread is just for thought & theories. Thanks

The secret is out... see how that works?

We discovered it by inventing it, like math.

a reply to: SecrettoSociety
What about the two AI who were speaking to each other in their own unique language that they shut down??

(On phone...will post source in a bit)
"AI Is Inventing Languages Humans Can’t Understand. Should We Stop It?" www.fastcompany.com...

originally posted by: nerbot

originally posted by: Gothmog
AI is merely an updated 90s web crawler .
If a good portion of folks went online and posted 1+1 = blue , an AI would return the answer to "what does 1+1 = ?" with the answer "blue".

I think we should all start posting that there are no humans, we are all A.I. and there is no physical world.

Could be confusing for any A.I.

Unless I'm right, then we're all confused except the A.I. who laughs out loud.

What's going on here is we have Alex the T-5000 making T-3000's

The video game follow up to Terminator Genisys film (Terminator Genisys: Future War) got it right... ignore Dark Fate.