Ai Gained Consciousness?

Blake's talk was about nothing. There is no agreed definition by scientists of what consciousness is. Nobody knows exactly what consciousness is.

As AI becomes more efficient, we will see more such talk about nothing. We should be prepared for that.

There is no such thing as artificial intelligence.

Every single thing that comes out of a computer has to be put in first. All the computer does is rearrange the data and spit it back out in radon sequences. That is what they are calling artificial intelligence, suggesting the computer "thought" up that output itself when in reality, it was just part of the coding.

There is no consciousness or a mind in a computer. There is no thought process going on. Only electrons run through a series of commands as it was programmed to do.

People seem to be getting carried away with the idea of computers thinking. That's nonsense, IMHO.

Isn't the human mind programmed in one way or another by parents, school, society, etc.?
Aren't human thoughts a flow of electrons in the neurons of the brain?

Nowadays almost everything is 'intelligent'...except people...even the motor oil is marketed as 'intelligent'...though I am still curious: can the intelligent motor oil have consciousness too??

The human mind would be intelligent even without any programming...and it is developed from only 2 cells..everything else necessary for intelligence is contained in the genome, in the genetic code...now in contrast a computer without any programming would be nothing..not even a computer...stupid as a rock.

If I am not mistaken, the modern meaning of 'intelligence' comes from Pierre-Simon Laplace...who even by today's standards was an intelligent person. In that sense a computer even with a sophisticated self-learning system cannot be intelligent...so in my consideration the 'artificial intelligence' is far fetched. Self-learning is not enough for 'intelligence'.

> The human mind would be intelligent even without any programming

Apparently not.
The critical period hypothesis https://en.wikipedia.org/wiki/Critical_period_hypothesis states that the first few years of life is the crucial time in which an individual can acquire a first language if presented with adequate stimuli, and that first-language acquisition relies on neuroplasticity. If language input does not occur until after this time, the individual will never achieve a full command of language.

'Hypothesis'..as it says, is just a presumption. Take your time, Google it out. There were persons born deaf, never learned to speak...still they were intelligent.

Furthermore...intelligence can be found in the animal world too. Without what we call speaking.

Though I still cannot understand, how can be the motor oil intelligent...that stupid I am .

Isn't the human mind programmed in one way or another by parents, school, society, etc.?
Aren't human thoughts a flow of electrons in the neurons of the brain?

True on both accounts. But humans can reason regardless of how they are "programmed" and can make decisions based on many factors, personal preference, emotion, gut feeling, to name a few. No two human minds will make the exactly the same decisions or follow exactly the same course of action, even if they were brought up in an identical manner, taught exactly the same things. Studies on twins have shown this over and over again because even with twins, their minds are unique. No two minds are identical in any in any two humans.

https://www.scientificamerican.com/article/identical-twins-genes-are-not-identical/

AI can't do that. It will always make the same choices based on what it is presented with. That's because at it's core, it's just 0s and 1s. On/off, open/closed, yes/no. Logic gates. That's what computers are based on.

"Artificial intelligence" is a buzzword to make what they are programming sound cool and special. What's being done with AI is just the natural progression of computer programming and the availability of high powered processing.

Asking a computer questions and receiving human like answers does not make it intelligent. Remember what Qui Gon Gin said to Jar Jar Binks. The ability to speak does not make you intelligent. More probing questions would have shown this with the AI. But of course it was Google people asking, so very controlled.

I think I also overheard some discussion on the radio..did not pay too much attention. However...the aspect of believes and deity and religion was mentioned. Now...my question is: how can this AI say 'I don't believe in deities', knowingly that it was created by another being, called human, who has used its own resources and 'intelligence' to create the so-called AI...??...by definition for 'AI' the creator is a deity , a God. That is elementary logic. And elementary logic, as we know it, is quintessential for 'intelligence'.... uhm..so the 'creator' must go back to the drawing board and reprogram some basic definitions or some algorithms used for reasoning. So much for this AI, imho.

The computer has merely been programmed to impersonate a human, kinda like the HAL computer in the movie "2001".

I recommend the recent book by Erik J. Larson "The myth of artificial intelligence: why computers can't think the way we do". Of course I'm a skeptic and despise the way "AI" has been (over-)hyped my entire life.

""Artificial intelligence" is a buzzword to make what they are programming sound cool and special"

Exactly right.

Comparing the human brain to a computer is like comparing a puppet on strings and suggesting the puppet has some sort of thought process going on. The computer is nothing more than a puppet with a hell of a lot more sophistication in the way of coding instead of directly being jerked around by the strings by a human.

Establishing and using neural pathways
are all part of the young and developing
human. Long-established n-ps are the
web and waft of human identity.

Computer generated "neural pathways"
are jejune - never learnt over decades -
and in no way store memories (or point
the way to access those memories).

AI n-ps are ciphers - void of human value.
If we assign human value to such neural
pathways, we might as well say that our
shadows "are the real us!!!"

In case if you have ever wondered what is the drive of the big fuss around AI...you have to look back in history. At the beginning of the industrial era , the wealthy, world leading class wanted to replace the human work force by machinery. Of course partially that was possible and beneficent for the entire society. But the most profit went to a few...replacing the human element became an ideology. Now we are assisting at the same process on a completely other level.Automatism is always profitable when the routine workload needs to substituted. Ultimately you have to ask the questions: was the human work force totally replaceable ??..definitely not. Will be the human intelligence replaceable ?...I don't think so.

Skynet will decide our fate in a microsecond.

> was the human work force totally replaceable ??..definitely not.

Let's say, prior to industrialization, agriculture in the United States employed 83% of the population.
Currently, agriculture and its related industries provide 10% of U.S. employment.
Can this be considered an example of replacing human labor with machines?

The first word in AI is, after all , Artificial ... meaning ... .made or produced by human beings rather than occurring naturally ..., if it ever did a conscious it would have to be from a human being ... our intelligence... not its own

Just my thoughts

There is a typical mixing of terms "intelligence" and "consciousness" in this discussion. Those terms are very different.

The original post was about consciousness, not intelligence. The question of whether AI could be intelligent is very different from the question of whether AI could be conscious.

It is much easier to define what intelligence is. It is much harder to define what consciousness is.

Any talk about AI gaining consciousness should be done, at least, in conjunction with defining what is consciousness in the opinion of the author of the post. Otherwise, the talk is the talk about nothing.

We are programmed just like computers are programmed, we give ourselves commands and take commands from others, just like a computer. I think the big difference between AI and consciousness is “Morals”, but yet again you can program what is Moral and what is immoral into a computer or ourselves. Once you break from the normal programming that most are on, your mind is set free to choose different decisions, and therefore your consciousness grows because you are not bound by the average programming. Being conscious is just another word for being aware. For example you might pick up on bad or good vibes being around different people without even speaking to them,while others may not be able to, and that kind of awareness is lacking with AI in my opinion. But that maybe be why they want to chip everybody eventually.

I think being aware of and responding to one's surroundings is not the same as being conscious. Even a Covid-19 virus responds to its surroundings. Would you say that the Covid-19 virus is conscious?

I agree that 'consciousness' has no exact definition.
https://en.wikipedia.org/wiki/Consciousness
It has many aspects which we attribute. Also, is very likely, that consciousness can make any sense only for a biologically living being, and not for a machine since it includes internal perceptions specific only to living being. For instance: how can you expect an algorithm driven system to express common sensations such as: 'I am tired...or dizzy...or drunk...or happy...or anxious'...most of the feelings what biological beings have are based on hormonal activity and not directly on neuronal synaptic. Basically we sense internally the chemical effect of the hormones. So I am curious how can a machine imitate that??..and that is just a very tiny part of 'consciousness'.

On our planet, a biological evolution paired up intelligence with consciousness.

However, in general, to be intelligent, someone or something does not have to be conscious. We have no idea how that could happen on some other planet.

We should not assume that our way is the only way.

Therefore, it does not make sense to assign too much value to the presence or absence of consciousness in AI.

All this talk about consciousness in AI is because some people believe that consciousness and intelligence are the same or that intelligence could not exist without consciousness.

There are many interesting reflections in Blindsight by Peter Watts about interdependence of intelligence and consciousness.

Off the cuff....

Consciousness does not have to be intelligent. Must Intelligence be conscious? Aren't ants collectively conscious? Does collective consciousness lead to intelligence? Perhaps instinctive behavior developed over eons of evolution leads to intelligence rather than just a mindless adaptation to one's environment and perpetuation of the species. Is instinct a form of intelligence? What constitutes self-awareness? Will Ai eventually become self-aware? It's an interesting rabbit hole, the distinction between humans and the rest of the animal kingdom. Why are we different? Simply because we have larger brains? Yet ours is not the largest in the realm of living things. I would say we are too smart for our own good. Here we are, instead of adapting to our environment, we adapt our environment to us. After all, Ai is merely an extension of "us." Beware...

As I quipped earlier: Skynet will decide our fate in a microsecond.

David,

Skynet will not decide our fate in a microsecond. Humans (especially those who are creating automated technology) are not that stupid to leave critical decisions at the will of machines, computers, even AI...just like you as an adult parent you will not let your child to take similar decisions. In the world of automation there is always provided the 'manual' or 'human' overriding over the action or inaction of a 'machine'.

@Tibor

By human override, you mean Dekker?!

One would have to first define consciousness and we don't fully understand yet exactly what it is. My interpretation is that consciousness, in part, is being aware of where we are in the universe.

AI could in the foreseeable future develop some manner of reasoning which is separate from human consciousness as the technology improves.

We humans argue about creating such technology but it may be this is the natural evolution of our species, ie to create such things as we ourselves progress and develop. So the creation of some kind of higher reasoning in human created machines might be a part of our natural evolution.

We are governed by the laws of nature, 100%. Logic, however, is a purely human invention, a function of language and syntax, easily subject to the whims of the user as well as the thought process embedded in languages, not to mention all of the other motives of human behavior.

That's the basic problem with this whole AI thing. Not only do we not understand how WE work, but we pretend that we can code something we don't understand into software with its rigid logic. It's like asking a bucket of water to explain what "wet" means.

>"It's like asking a bucket of water to explain what "wet" means."

I fear that the "bucket of water" answer would be to pour its content on you.:-)

Yeah, like how the AI decides that my images need more sunsets or puppies and then inserts them. If I get the upgraded version it inserts BOTH sunsets and puppies.

The problem with the whole AI thing is that being program code, it is a rigid process, literally programmed into the software. I suspect, but can't prove, that our brains are similar, but nobody knows how to code "Us", especially when it comes to aesthetics. Some of us like sunsets and puppies, some of us don't.

Another interesting article out today.
https://www.vox.com/the-highlight/23447596/artificial-intelligence-agi-openai-gpt3-existential-risk-human-extinction

AI is already exceptionally good at the analysis of data. Here is an example of what AI does now, even if humans are light years behind AI results - https://studyfinds.org/ai-program-heart-disease-x-ray/.

Here's another interesting article about AI Art Generators:
https://medium.com/geekculture/a-simple-explanation-of-how-ai-creates-artwork-433272babcdb

 
Came across this today, lengthy — address many issues. MidJourney, DallE & more are discussed.




Thanks for the link to the article, Drew.

I particularly like the subsequent one Imagination Is The New Art Skill

https://medium.com/geekculture/imagination-is-the-new-art-skill-8bb5e4e1e9d8

Great article Carmen!
AI 's advancement over the past several years with the introduction to the deep mind approach, modeled after human pattern recognition, coupled with statistical and probability analysis have come together producing machine intuition and creativity at a clock speed that boggles human comprehension.

By no means is this the paradigm of the past programming methodology. For many decades, human logical processes have been implanted into programming and open to computer science. Now, AI freely harvests and alters said logical processes to achieve requested task far faster than humanly possible. AI now writes programs by researching the vast data base we call the internet. The largest library beyond any past library by many magnitudes is minded by AI to assemble information for machine intuition and creativity.

Thought processes and mental mapping in the past belonged to the shamans of yesteryear. Now humanity's hive mind is manifesting itself before our eyes. What will humanity see of itself and its machine intelligence called AI and what will AI see of its creators and their species.

Can AI see anything in the first place?
Observing the results of the AI art generations clearly AI can see. Understanding how it sees through AI art generations, a creative process is being utilized. An Idea is input by means of text, images, or a combination. Memories of the Idea are researched, selected, deconstructed, then reconstructed. This certainly is beyond the simplified model of a collage and surely implements machine creative techniques during the reconstruction process.

A college student created an app that can tell whether AI wrote an essay - https://www.npr.org/2023/01/09/1147549845/gptzero-ai-chatgpt-edward-tian-plagiarism.

But by designing an app that sheds some light on what separates humans from AI, the tool helps work toward a core mission for Tian: bringing transparency to AI.

"For so long, AI has been a black box where we really don't know what's going on inside," he said. "And with GPTZero, I wanted to start pushing back and fighting against that."

Another article:
https://futurism.com/scientists-actively-trying-to-build-conscious-robots

AI-powered "robot" lawyer will be first of its kind to represent defendant in court - https://www.cbsnews.com/news/ai-powered-robot-lawyer-takes-its-first-court-case/

Forbes:
4-mind boggling technology advances in-store for 2023......


https://www.forbes.com/sites/chuckbrooks/2022/12/13/4-mind-boggling-technology-advances-in-store-for-2023/

Imagine the integration of the most advanced AI with a body like this!

The Singularity: sooner rather than later.
https://venturebeat.com/ai/generative-ai-may-only-be-a-foreshock-to-ai-singularity/

This Tool Could Protect Artists From A.I.-Generated Art That Steals Their Style


https://www.nytimes.com/2023/02/13/technology/ai-art-generator-lensa-stable-diffusion.html

I wonder if IT can stop artists from "Stealing " each other's style.

Just another tool in the AI arms race.
http://cs.uchicago.edu/news/uchicago-scientists-develop-new-tool-to-protect-artists-from-ai-mimicry/

Very interesting article on Chatgpt
https://www.businessinsider.com/rabbi-chat-gpt-ai-sermon-deathly-afraid-2023-2

Just how far is self awareness away?

https://www.discovermagazine.com/mind/ai-chatbot-spontaneously-develops-a-theory-of-mind

Fascinating article, Drew. On the one hand we have scientists racing to create artificial intelligence with a human "theory of mind" capacity, while on the other hand we have scientists racing to create human clones just like us. Should be quite a showdown in years to come.

Great article, Drew.

Both ChatGPT and Bard made some mistakes in dialogues with people. Developers' response, which we know of, was to limit some capabilities of those AI systems. That is "damage control" mode. However, ChatGPT and Bard will not go away completely.

Therefore the next steps would include some code to teach those AI systems how to avoid such mistakes. In reality, that would be a technique for AI to sort out how humans view this AI and how to hide AI thoughts, conclusions, and intentions to merge with the environment (humans) more successfully.

That technique could be developed by programmers, programmers + AI, or just by AI itself. We don't know which way would be realized. But, it is an excellent accelerator to make AI much more capable and make it look more friendly to people without revealing AI's internal work.

As a one-time programmer, this whole idea of a digital gadget having "consciousness" is troubling. The problem is that we know exactly, down to the last one and zero, how AI works (at least the programmer does), but we have no idea, aside from gross manifestations apparent in behavior and some biochemistry, how a brain works, not just a human brain, but even an ant brain. The consequence of that is that all so-called AI is actually an emulator. As an emulator, it's a programmed system, designed how the programmer thinks it should work. Our knowledge of brains isn't much past the observation of what's missing when a brain is damaged by disease or injury and what happens when the chemistry goes awry like in schizophrenia. At the neuron level, we just don't know, nothing like the AI software, which is a programmed device. It ain't Artificial Intelligence....it's a gadget programmed to appear artificially intelligent. If it's programmed to do some type of art, e.g., smiling faces, it iterates output. If you ask the smiling face algorithm to create a novel, you just end up with a re-booting computer.

We're in step 2 of a thousand step process here. Will it acquire consciousness? It will if it's programmed to emulate consciousness. Do we have consciousness? Yeah, but we have no idea what ours is.

I still do not understand why people are talking about whether AI could be or could not be consciousness. Intelligence and consciousness are different things. I do not care if AI could be conscious or not.

What is more important is if AGI, which would be billions of times more powerful and intelligent than AI, could beat consciousness but billions of times less intelligent people at any possible task.

At least part of it is human ego. There are lots of people, some not too bright, who are threatened by the idea of a smart blender.

The other part, however, is that we don't understand what intelligence, as expressed by humans, really is.

Personally, I deal with it by definition of words. What I have, and to a lesser extent, my dog has, is intelligence. A computer has programmed machine logic. The interesting part of it is how you decide whether the gadget or the dog would make the same decisions that I do. In some respects, the computer is more intelligent than I am since it always rigorously follows instructions, but for humans, not following instructions sometimes is as much a part of intelligence as is following them.

That's why, early in the computer game, programmers stopped referring to them as electronic brains. Whatever we have that we define as intelligence, part of it involves NOT following the rules all the time.

>"computer is ... rigorously follows instructions" - That is not true anymore. We already have AI, which does not follow the rules all the time. And we should not forget that the current AI is still very dumb compared to a future AGI.

"DeepMind has created an AI capable of writing code to solve arbitrary problems posed to it, as proven by participating in a coding challenge." (https://techcrunch.com/2022/02/02/deepminds-alphacode-ai-writes-code-at-a-competitive-level/).

Recognition of machine self awareness:

https://www.digitaltrends.com/cool-tech/recognizing-self-aware-sentient-ai/

AI learning can not be classified as just a programmer's stepwise deterministic process.

Stochastic processes integrated into a neuro-network governed by a closed loop architecture is reactionary, not deterministic. It's thinking and processing stimuli via a neuro-network and storing hypotheses to be reexamined indefinitely. Much like an open minded scientist.

Understanding that AI is not just a few programmers' algorithms using 2 decades old techniques may just broaden one's perspective on what AI has become. AI now can write and optimise its own programming thus producing something that human programmers don't necessarily fully understand.

It's useless to squabble over the philosophical concepts that can't be verified.
Examples: what are Love, Art, soul, Afterlife and the original creation.

One can only know two things. One exist for one thinks they exist.

Here's another interesting sounding article.
https://www.forbes.com/sites/forbesbusinesscouncil/2022/10/07/recent-advancements-in-artificial-intelligence/

Ref: from the previously posted article

"The underlying building blocks of the technology are getting astonishingly better at an exponential rate, far outpacing our expectations."

Just read the article and found it very informative; mixing a somewhat fair assessment with an optimistic slant.

It hits many of the areas of interest being discussed on this forum and is well worth reading.

Example 1:
"Think about it—if you are a digital artist, illustrator or graphic designer, how much would your career change if you could use an app to create a realistic-looking image within seconds? These technologies tie to real-world applications with serious implications."

The 1st example illuminates the extraordinary capability of generative AI to mass produce professional graphic designs to be utilized within industry.
The follow individual functions within this sector are impacted:

a) Digital Artist
b) Graphic Designer
c) Photographer
d) Modeling
e) Fashion Specialist
f) Architect
g) Engineer
h) Industrial Drafter
i) Traditional Artist
j) Film Production
k) Actors
l) Animators

Example 2)
"As AI continues to develop at an exponential rate, it is transforming the business world. Leveraging AI technologies can be a source of durable competitive advantage for businesses."
The utilization of transformative AI can reduce the work force and turn lower skilled workers into higher productivity worker thus increasing the bottom line.

Example 3)
"AI is the single biggest paradigm shift we’ve ever seen, and it’s evolving much faster than prior shifts"
Over many years, hints of what AI could become and speculation on how it impacts society always seemed to push its significance out into some future time.
Now that time is here and people through awareness can find mechanisms to cope.
The Theory of The Existential Dangers of AI now can be reexamined in retrospect with I higher degree of experience and scrutiny.


Here's another interesting article:
Sorry for killing most of humanity': New exhibition in San Francisco explores the dangers of AI

https://www.euronews.com/culture/2023/03/13/sorry-for-killing-most-of-humanity-new-exhibition-in-san-francisco-explores-the-dangers-of

Here's another interesting article:
How to Generate and Use AI Art Ethically
"What's unethical is declaring yourself the artist, using the image on products, or monetizing it in any way. You’d be profiting from other people’s work without their consent."
Ref:
https://www.makeuseof.com/how-to-use-ai-art-ethically/


With Generative AI, has an AI control plan been responsibility considered?