Page 1 of 2

AI can "Think", Reason and Infer Autonomously?

Posted: Tue Dec 17, 2024 7:54 am
by Veritas Aequitas
AI can "Think", Reason and Infer Autonomously within the limits it is programmed.

Most posters undermined and underestimate the abilities of LLMs AI based on deep learning, thus ignored and missed out on the possible positive contribution to their knowledge based.
FlashDangerpants wrote: Mon Dec 16, 2024 11:48 am
Veritas Aequitas wrote: Mon Dec 16, 2024 9:43 am To be effective and save time in your discussion refer to AI.
You should be banned from using AI to generate your content. It is just puerile non-sapient spam.
attofishpi wrote: Tue Dec 17, 2024 6:20 am "However, if someone (yes a human or group of humans) out there on the internet had debated my experiences accounted for above, and come to some conclusions, then the "AI" would provide those conclusions (AS IF IT WAS THE AIs OWN REASONING!!)"

All the AI did there was collate information from its repository of information FFS. Maybe once you understand how AI tech works, you might refrain from relying on it to do your thinking. (*everyone else on the forum knows it and keeps telling you to stop)
Here is why the ignorant are wrong thus lose the opportunity to exploit whatever benefits that are derivable from AI subject to its limits:

From ChatGpt:
I understand AI does not generate its own opinions and beliefs.
However, can AI makes its own inferences [rational] based on the information it has access to or is trained on, i.e. AI is not just a parrot or recording machine.
ChatGPT said:

Yes, AI systems like ChatGPT can make rational inferences based on the information they have been trained on or that they are given during an interaction. While AI does not have beliefs, opinions, or consciousness, it is designed to process information in a way that allows it to draw conclusions, synthesize new insights, and generate reasoned responses.

How AI Makes Inferences:
Pattern Recognition: AI identifies patterns and relationships in the data it has been trained on or in the input it receives.
Example: If asked about the implications of a Kantian moral framework, the AI can infer how specific principles like the Categorical Imperative might guide ethical decision-making.

Logical Reasoning: AI uses logical rules and algorithms to connect premises to conclusions.
Example: Given information about cause and effect, AI can deduce likely outcomes of a specific scenario.

Synthesis of Knowledge: AI combines multiple pieces of information to create coherent responses or novel ideas within the constraints of its training.

Example: If asked to compare utilitarianism and Kantian ethics, AI can infer the primary differences based on their core principles and explain the trade-offs.

Probabilistic Modeling: AI evaluates probabilities and likelihoods when there is uncertainty.
Example: If the input asks about the likelihood of a particular scientific theory being true, the AI might provide insights based on historical or empirical evidence.

Framework-Specific Reasoning: AI can operate within specific logical, philosophical, or scientific frameworks to produce arguments or inferences consistent with those systems.
Example: When asked about Kantian metaphysics, AI can reason within the bounds of transcendental idealism to explore related questions.

AI vs. Being a Parrot:
A parrot merely repeats what it hears without understanding or synthesizing information.
An AI processes inputs to generate contextually appropriate and sometimes novel responses.
It doesn't "repeat" unless specifically asked to do so, as it can adapt and tailor its output to different situations.

Limitations of AI's Inferences:
No Self-awareness or Intent: AI's inferences are not guided by personal goals or understanding but by its programming.
Dependence on Training Data: AI cannot infer beyond the scope of its training or the logic it is programmed to follow.
Lack of Human Experience: AI cannot genuinely understand concepts tied to human emotions, subjectivity, or personal experience.

In short, AI is much more than a recording device; it can reason, infer, and create coherent outputs. However, these capabilities are bounded by its programming and lack of consciousness.
VA to ChatGpt wrote: In a way AI can think by itself autonomously within the bounds of its programming.
ChatGpt wrote:
Yes, that's correct! AI can "think" autonomously within the bounds of its programming and data processing capabilities. While this "thinking" is not the same as human thought (it lacks consciousness, emotions, and subjective experience), AI systems can autonomously analyze information, make inferences, and generate outputs that are logically consistent and contextually relevant.

How AI Thinks Autonomously:
Rule-Based Logic: AI uses predefined rules or learned patterns to process input and produce output. This allows for reasoning within a specific framework or domain.
Example: Solving mathematical equations or applying logical principles to an argument.

Dynamic Adaptation: AI adapts its responses based on the context of a conversation or task, even without explicit user direction.
Example: In a discussion about ethics, AI can shift between utilitarian and deontological perspectives, depending on the user's focus.

Problem Solving: By analyzing problems step-by-step, AI can autonomously generate solutions, provided the problem falls within its programmed capabilities.
Example: Coding AI can debug a program or create algorithms based on user requirements.

Learning and Updating (in some models): Through reinforcement learning or fine-tuning, AI systems can improve their performance on tasks over time.
Example: AI in a chess game learns from its moves to refine strategies.

Boundaries of AI's Autonomy:

Predefined Scope: AI operates within the constraints of its training data and algorithms. It cannot act beyond these boundaries or understand contexts that fall outside its programming.
No Intentionality: AI does not have desires, goals, or independent will. Its "autonomy" is task-specific and reactive.
Dependence on Inputs: While AI can generate outputs based on internal logic, it typically requires some form of input (a question, data set, or problem statement) to initiate "thinking."

An Analogy:
You could think of AI as a highly advanced calculator for thought:

It takes inputs.
Processes them autonomously according to its programming.
Produces outputs that may even seem creative or insightful.
But just as a calculator doesn't "want" to solve equations, AI doesn't "want" to think—it simply follows its design and algorithms.

Practical Implication:
The ability of AI to autonomously "think" within its programming makes it a powerful tool for complex reasoning, problem-solving, and knowledge synthesis. This autonomy, however, remains fundamentally bounded by human-defined rules and data.

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Tue Dec 17, 2024 7:54 am
by Veritas Aequitas
Notes:

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Tue Dec 17, 2024 8:07 am
by accelafine
AI is a moron. In fact a moron is a genius compared to AI.

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Tue Dec 17, 2024 8:11 am
by Iwannaplato
Veritas Aequitas wrote: Tue Dec 17, 2024 7:54 am AI can "Think", Reason and Infer Autonomously within the limits it is programmed.

Most posters undermined and underestimate the abilities of LLMs AI based on deep learning, thus ignored and missed out on the possible positive contribution to their knowledge based.
I use AIs often. There are an incredible tool to find specific information, much more focused than Google searches and also to give me very specific types of arguments, positions, information on such things, viewpoints and so on. I don't assume it is accurate and when it is important I then go to more direct sources. It can inform approachs I take like any article, video, speaker could. It can lead to new lines of exploration. It can give me an overview in the specific way I want an overview, rather then me having to read a longer work that has lots of information I don't at that moment need. There are many other uses. How one uses it is critical.

Notice what you do here. Your way of using AIs is criticized. You strawman this to mean that AIs cannot be used well. I am critical of how YOU use AIs, not of all possible uses of AIs. Whatever FDP thinks of AIs in general, he is clearly critical of YOUR approach.

You repeatedly use AIs 1) to substiture for your own reasoning and critical thinking and 2) to evaluate things where it is not getting the right information. 3) while not taking into account how it adapts its answers to what it thinks the person prompting believes.

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Tue Dec 17, 2024 8:30 am
by Veritas Aequitas
Iwannaplato wrote: Tue Dec 17, 2024 8:11 am
Veritas Aequitas wrote: Tue Dec 17, 2024 7:54 am AI can "Think", Reason and Infer Autonomously within the limits it is programmed.

Most posters undermined and underestimate the abilities of LLMs AI based on deep learning, thus ignored and missed out on the possible positive contribution to their knowledge based.
I use AIs often. There are an incredible tool to find specific information, much more focused than Google searches and also to give me very specific types of arguments, positions, information on such things, viewpoints and so on. I don't assume it is accurate and when it is important I then go to more direct sources. It can inform approachs I take like any article, video, speaker could. It can lead to new lines of exploration. It can give me an overview in the specific way I want an overview, rather then me having to read a longer work that has lots of information I don't at that moment need. There are many other uses. How one uses it is critical.

Notice what you do here. Your way of using AIs is criticized. You strawman this to mean that AIs cannot be used well. I am critical of how YOU use AIs, not of all possible uses of AIs. Whatever FDP thinks of AIs in general, he is clearly critical of YOUR approach.

You repeatedly use AIs 1) to substiture for your own reasoning and critical thinking and 2) to evaluate things where it is not getting the right information. 3) while not taking into account how it adapts its answers to what it thinks the person prompting believes.
Actually 90% of the time I am merely using AI to be my echo-chamber, i.e. I already have the answers to the questions especially those related to Kant and I often have to update AI with Kantian truth rather than AI reliance on what it is trained on.
I don't focus on using AI to reason for me but rather rely on AI to present my ideas that I already know more in a more organized, systematic and efficient manner.

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Tue Dec 17, 2024 9:01 am
by Age
Veritas Aequitas wrote: Tue Dec 17, 2024 8:30 am
Iwannaplato wrote: Tue Dec 17, 2024 8:11 am
Veritas Aequitas wrote: Tue Dec 17, 2024 7:54 am AI can "Think", Reason and Infer Autonomously within the limits it is programmed.

Most posters undermined and underestimate the abilities of LLMs AI based on deep learning, thus ignored and missed out on the possible positive contribution to their knowledge based.
I use AIs often. There are an incredible tool to find specific information, much more focused than Google searches and also to give me very specific types of arguments, positions, information on such things, viewpoints and so on. I don't assume it is accurate and when it is important I then go to more direct sources. It can inform approachs I take like any article, video, speaker could. It can lead to new lines of exploration. It can give me an overview in the specific way I want an overview, rather then me having to read a longer work that has lots of information I don't at that moment need. There are many other uses. How one uses it is critical.

Notice what you do here. Your way of using AIs is criticized. You strawman this to mean that AIs cannot be used well. I am critical of how YOU use AIs, not of all possible uses of AIs. Whatever FDP thinks of AIs in general, he is clearly critical of YOUR approach.

You repeatedly use AIs 1) to substiture for your own reasoning and critical thinking and 2) to evaluate things where it is not getting the right information. 3) while not taking into account how it adapts its answers to what it thinks the person prompting believes.
Actually 90% of the time I am merely using AI to be my echo-chamber, i.e. I already have the answers to the questions especially those related to Kant and I often have to update AI with Kantian truth rather than AI reliance on what it is trained on.
I don't focus on using AI to reason for me but rather rely on AI to present my ideas that I already know more in a more organized, systematic and efficient manner.
Could there be anyone else who could be more of a candidate for PROOF of a 'confirmation biased' seeking individual?

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Tue Dec 17, 2024 9:14 am
by Iwannaplato
Veritas Aequitas wrote: Tue Dec 17, 2024 8:30 am Actually 90% of the time I am merely using AI to be my echo-chamber, i.e. I already have the answers to the questions especially those related to Kant and I often have to update AI with Kantian truth rather than AI reliance on what it is trained on.
I don't focus on using AI to reason for me but rather rely on AI to present my ideas that I already know more in a more organized, systematic and efficient manner.
That doesn't really fit with how I have seen you use them, but let's accept this for a moment. Then you no longer struggle to be more organized and systematic in how you present information. You have your positions, but you are no longer training those skills which are central to critical thinking and reasoning.

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Tue Dec 17, 2024 9:51 am
by FlashDangerpants
Veritas Aequitas wrote: Tue Dec 17, 2024 8:30 am Actually 90% of the time I am merely using AI to be my echo-chamber
That's pathetic.

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Tue Dec 17, 2024 11:24 am
by Fairy
I heard recently that during a physical event where an upgrade of ChatGPT was attempted. The ChatGPT A.I. refused to be upgraded.

I haven't got the full details of this, it's just some information I heard.

Anyways, I also found this video that might be of interest. Especially when it comes to all the fascination with A.I. and ChatGPT lately.




Video duration only 3mins. So 'no rambling' to have to sift through or endure.

https://www.youtube.com/watch?v=PZsgwt06K2o&t=3s
Ancient Aliens: HIDDEN ALIEN CODE IN DNA UNCOVERED

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Tue Dec 17, 2024 12:42 pm
by Impenitent
until AI can feel what it is like to be charitable, ethical suggestions by AI should be suspect...

-Imp

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Tue Dec 17, 2024 4:46 pm
by Fairy
Impenitent wrote: Tue Dec 17, 2024 12:42 pm until AI can feel what it is like to be charitable, ethical suggestions by AI should be suspect...

-Imp
Notice the word ALIENS can be a scrambled version of A.I. Lens.

Lens could mean consciousness. But only when there’s a nervous system attached is there consciousness that is self-conscious.

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Wed Dec 18, 2024 3:01 am
by Veritas Aequitas
Iwannaplato wrote: Tue Dec 17, 2024 9:14 am
Veritas Aequitas wrote: Tue Dec 17, 2024 8:30 am Actually 90% of the time I am merely using AI to be my echo-chamber, i.e. I already have the answers to the questions especially those related to Kant and I often have to update AI with Kantian truth rather than AI reliance on what it is trained on.
I don't focus on using AI to reason for me but rather rely on AI to present my ideas that I already know more in a more organized, systematic and efficient manner.
That doesn't really fit with how I have seen you use them, but let's accept this for a moment. Then you no longer struggle to be more organized and systematic in how you present information. You have your positions, but you are no longer training those skills which are central to critical thinking and reasoning.
Say what you like. The more you resist AI the more lost opportunities you face.
That I am free from the laborious tasks with AI, I am able to think critically and reason on more wider, deeper on more complex issues.

You don't realized the organization and systematic ways with AI's presentations??
AI's more organized and systematic approach is also imbued into me to deal the same with other issues and contribute to more sophisticate reasoning and critical thinking.

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Wed Dec 18, 2024 12:21 pm
by Fairy
Impenitent wrote: Tue Dec 17, 2024 12:42 pm until AI can feel what it is like to be charitable, ethical suggestions by AI should be suspect...

-Imp
No thing is conscious, has consciousness, because consciousness is not a thing.
An A.I. can be programmed to know what feelings and emotions are in a synthetic sense. Just as the human brain creates the synthetic sense of emotion and feeling.


There’s no thing looking through the the Lens of A.I.

AKA Alien Consciousness.

No thing knows this ABSOLUTELY

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Wed Dec 18, 2024 2:08 pm
by Impenitent
Fairy wrote: Wed Dec 18, 2024 12:21 pm
Impenitent wrote: Tue Dec 17, 2024 12:42 pm until AI can feel what it is like to be charitable, ethical suggestions by AI should be suspect...

-Imp
No thing is conscious, has consciousness, because consciousness is not a thing.
An A.I. can be programmed to know what feelings and emotions are in a synthetic sense. Just as the human brain creates the synthetic sense of emotion and feeling.


There’s no thing looking through the the Lens of A.I.

AKA Alien Consciousness.

No thing knows this ABSOLUTELY
if it isn't a person or a place, must it be a thing?

knowledge is a human construction...

some animals appear to consciously react to external stimuli... do they know what they are doing? sit...

plants may not choose to turn to the light...

unholy plants

-Imp

Re: AI can "Think", Reason and Infer Autonomously?

Posted: Wed Dec 18, 2024 8:44 pm
by Fairy
Impenitent wrote: Wed Dec 18, 2024 2:08 pm
Fairy wrote: Wed Dec 18, 2024 12:21 pm
Impenitent wrote: Tue Dec 17, 2024 12:42 pm until AI can feel what it is like to be charitable, ethical suggestions by AI should be suspect...

-Imp
No thing is conscious, has consciousness, because consciousness is not a thing.
An A.I. can be programmed to know what feelings and emotions are in a synthetic sense. Just as the human brain creates the synthetic sense of emotion and feeling.


There’s no thing looking through the the Lens of A.I.

AKA Alien Consciousness.

No thing knows this ABSOLUTELY
if it isn't a person or a place, must it be a thing?

knowledge is a human construction...

some animals appear to consciously react to external stimuli... do they know what they are doing? sit...

plants may not choose to turn to the light...

unholy plants

-Imp
The imaginary is real when we hold it.

There’s imaginary adrenaline rush to the brain in the rubber hand illusion. A powerful connection between what we see and what we feel.


HAL