Why is The Gettier problem still considered an open issue?

[Age]

My belief that there is a television in my living room right now is fully justified by
seeing a television in my living room right now.

This remains true if I am a brain in a vat, hooked up to a computer. When-so-ever
a set of what appears to be physical sensations correctly maps to the elements of
the model of the world we have truth and knowledge.

This here is Wrong and Incorrect because of the use of the 'appear' word here.

Alternatively we could disavow all empirical knowledge and call this conjecture instead.
This only leave analytical knowledge. Analytical knowledge is a set of mutually self-defining
semantic tautologies. https://www.britannica.com/topic/tautology

A statement IS 'True' IF PROOF for 'that statement' IS KNOWN.

If PROOF for a statement IS NOT YET KNOW, then 'that statement' COULD BE 'true'. That is; if PROOF for 'that statement' being 'False' IS NOT KNOWN.

In other words, ANY statement can only be Truly JUSTIFIED WITH ACTUAL PROOF.

'BELIEF' of or for absolutely ANY 'thing' is completely REDUNDANT here, and is BEST left OUT OF 'the equation', altogether.

If we really are brains in vats then the belief that physical reality exists is false.

'We' REALLY are NOT 'brains in vats', SO MOOT.

Also, even if 'we' REALLY were 'brains in vats', then this does NOT mean that 'physical reality exists if false' AT ALL. In fact if 'we' REALLY were 'brains in vats', then this ACTUALLY MEANS that 'physical reality' ACTUALLY DOES EXIST.

None-the-less {cats} <are> {animals} in the axiomatic model of the world.

In WHOSE 'axiomatic model' of 'the world'?

WORK 'this' OUT, and then you WILL BE CLOSER to SEEING and UNDERSTANDING what the ACTUAL and IRREFUTABLE Truth of 'things' REALLY IS.

[Age]

THEN JUST define 'the (so-called) nature of knowledge', PRECISELY.

Actual knowledge is a semantic tautology containing within the axiomatic system
of the model of the world. Knowledge is expressed as sound deductive inference.

SO WHY THEN are you 'TRYING TO' I precisely
define the nature of knowledge?

The correspondence between (what at least appears to be) physical sensations

What do you mean by 'at least APPEARS to be' 'physical sensations'?

Do you FEEL 'physical sensations' or NOT?

If you DO, then what was the 'appears to be' words USED here for, EXACTLY?

And, if you DO FEEL 'physical sensations', then will you provide us with some examples?

and this model of the world is at best conjecture that at least seems to be very
reliable.

So, if what APPEARS TO BE 'physical sensations', to you, and the correspondence between WHAT APPEARS TO BE 'physical sensations' and 'this model' (whatever model that IS, EXACTLY,) of 'the world' is at best conjecture that at least SEEMS TO BE VERY RELIABLE, to you, then WHAT is the ACTUAL ISSUE here, EXACTLY?

[Age]

It is everything known to man that can be expressed using language

Not empirical conclusions. The AI can't draw any conclusions about the world directly. It has no situation dependant knowledge. Pretty much all the things we use at work, play, problem solving.

It can more or less define things and check if definitions make sense or paraphrases make sense. That's about it.

But what's the purpose of the AI in question? What are its intended functions?

Situation dependent knowledge is provided as input.
It is to become capable of anything that a human
mind can do at the expert level of each field.

There is NO ACTUAL SUCH 'thing' as the MISNOMER 'human mind'.

Therefore, it is an IMPOSSIBILITY for absolutely ANY 'thing' to become capable of what an UNKNOWN 'thing' is IMAGINED to be capable of doing.

Artificial intelligence, just like the human brain, which, by the way, works EXACTLY like a computer does, can NOT DO ANY 'thing' that has NOT been PREVIOUSLY INPUTTED INTO 'it'.

Artificial intelligence can NOT do what is done at the so-called 'expert level of each field'. Artificial intelligence just COPIES what has ALREADY BEEN FOUND, DESIGNED, and/or CREATED.

Just like the human brain can and does ONLY process with 'what' has ALREADY BEEN FED INTO 'it' or is AVAILABLE TO 'it'.

I am not getting notified when you post.

[Agent Smith]

Not empirical conclusions. The AI can't draw any conclusions about the world directly. It has no situation dependant knowledge. Pretty much all the things we use at work, play, problem solving.

It can more or less define things and check if definitions make sense or paraphrases make sense. That's about it.

But what's the purpose of the AI in question? What are its intended functions?

Situation dependent knowledge is provided as input.
It is to become capable of anything that a human
mind can do at the expert level of each field.

There is NO ACTUAL SUCH 'thing' as the MISNOMER 'human mind'.

Therefore, it is an IMPOSSIBILITY for absolutely ANY 'thing' to become capable of what an UNKNOWN 'thing' is IMAGINED to be capable of doing.

Artificial intelligence, just like the human brain, which, by the way, works EXACTLY like a computer does, can NOT DO ANY 'thing' that has NOT been PREVIOUSLY INPUTTED INTO 'it'.

Artificial intelligence can NOT do what is done at the so-called 'expert level of each field'. Artificial intelligence just COPIES what has ALREADY BEEN FOUND, DESIGNED, and/or CREATED.

Just like the human brain can and does ONLY process with 'what' has ALREADY BEEN FED INTO 'it' or is AVAILABLE TO 'it'.

I am not getting notified when you post.

Stop, stop, stop!

Now here, ladies, gents, is a fine specimen of a philosopher! According to Agent Smith of course, a long-time staff in our museum of curiosities. It will come as a great relief to you that he wasn't fired, that happened yesterday, for lack of skill but for, ahem, sexually deviant behavior.

[Skepdick]

That it all in its axiomatic model of the world.

Olcott, you are a fucking idiot.

You are using an axiomatic system (Mathematics) to model a non-axiomatic system (reality).

Kant’s basic argument is that mathematicians are justified in constructing objects or axioms a priori, because they can work with pure intuitions—like a line or the form of a triangle, for example—rather than being restricted only to the analysis of concepts. (See the entry on Kant’s philosophy of mathematics.) Philosophers cannot follow this sort of procedure because they have no right to assume any a priori intuitions or axioms about metaphysical entities. Attempts that rely on such claims have only produced “so many houses of cards” (A727/B755).

https://plato.stanford.edu/entries/kant ... aCogRolLim

[Iwannaplato]

Situation dependent knowledge is provided as input.
It is to become capable of anything that a human
mind can do at the expert level of each field.

Wouldn't this need the ability to use empirical conclusions? and also to draw them?

I am not getting notified when you post.

Odd. I'm quoting from your posts to create responses like I always do. Unless everyone else has kept quiet about my messing up for years about this, It must be a glitch exception.

That it all in its axiomatic model of the world.

I don't understand that sentence or what it's referring to. If you mean the AI you are designing will have an axiomatic model of the whole world. Perhaps in some sense. But it can't produce any new information. It can't experiment or create. It can simple check verbal and mathematical assertions to see if they fit the mathematical rules and definitions it has. I think we have AIs that can do this already. But I am no expert, not close.
What could this AI do with its math abilities and abilities to check definitions and semantics?

[Agent Smith]

Why is the elephant ... in the room, why is the bull ... in the China shop, why is the canary ... in the coal mine, why is the dog ... in the manger and why, why, why is the fish ... out of water? Why? Why? WHY?!!

[PeteOlcott]

Wouldn't this need the ability to use empirical conclusions? and also to draw them?


Odd. I'm quoting from your posts to create responses like I always do. Unless everyone else has kept quiet about my messing up for years about this, It must be a glitch exception.

That it all in its axiomatic model of the world.

I don't understand that sentence or what it's referring to. If you mean the AI you are designing will have an axiomatic model of the whole world. Perhaps in some sense. But it can't produce any new information. It can't experiment or create. It can simple check verbal and mathematical assertions to see if they fit the mathematical rules and definitions it has. I think we have AIs that can do this already. But I am no expert, not close.
What could this AI do with its math abilities and abilities to check definitions and semantics?

When an AI model has all of the general knowledge of the world that every human expert has in every field
and it also has all of the details of common sense that humans have it will have the functional equivalent of
a human mind. We also have divided analytic and synthetic as everything that it knows is analytic and the only
things that it does not know are synthetic.

In any case The Gettier problem is refuted. When-so-ever a proposition is fully justified as true such that this justification necessitates its truth it affirms the proposition as knowledge.

In those cases where the justification is stochastic rather then deterministic we have reasonably plausible conjecture rather than knowledge. The boundary condition here is the knowledge of specific instances of reasonably plausible conjecture. It will know that it does not know X and will also know that X is the closest that anyone currently has to knowledge of Y.

My model of the world as the analytic side of the analytic synthetic distinction seems to address Quine's holism. In philosophy of science, confirmation holism, also called epistemological holism, is the view that no individual statement can be confirmed or disconfirmed by an empirical test, but rather that only a set of statements (a whole theory) can be so. It is attributed to Willard Van Orman Quine.
https://en.wikipedia.org/wiki/Confirmation_holism

[PeteOlcott]

So, if what APPEARS TO BE 'physical sensations', to you, and the correspondence between WHAT APPEARS TO BE 'physical sensations' and 'this model' (whatever model that IS, EXACTLY,) of 'the world' is at best conjecture that at least SEEMS TO BE VERY RELIABLE, to you, then WHAT is the ACTUAL ISSUE here, EXACTLY?

The Gettier problem is easily abolished by defining knowledge as the subset of
justified true beliefs such that the justification guarantees the truth of the belief.

Those cases where the justification does not guarantee the truth of the belief,
yet provides some stochastic measure that the justification makes the belief very
plausible or highly likely are conjectures and not knowledge.

[PeteOlcott]

Artificial intelligence, just like the human brain, which, by the way, works EXACTLY like a computer does, can NOT DO ANY 'thing' that has NOT been PREVIOUSLY INPUTTED INTO 'it'.

Artificial intelligence can NOT do what is done at the so-called 'expert level of each field'. Artificial intelligence just COPIES what has ALREADY BEEN FOUND, DESIGNED, and/or CREATED.

Just like the human brain can and does ONLY process with 'what' has ALREADY BEEN FED INTO 'it' or is AVAILABLE TO 'it'.

That is has been refuted empirically
https://hai.stanford.edu/news/examining ... age-models
There are numerous concrete examples where a LLM solved a problem in a way that
no human ever thought of. In addition to this these LLM gain many abilities that they
were not trained for. ChatGPT began to process all information much more logically
when it was taught about computer programming.

[Iwannaplato]

When an AI model has all of the general knowledge of the world that every human expert has in every field
and it also has all of the details of common sense that humans have it will have the functional equivalent of
a human mind.

So, both empirical and non-empirical knowledge.

We also have divided analytic and synthetic as everything that it knows is analytic and the only
things that it does not know are synthetic.

So, how will it treat empirical knowledge differently from how experts treat it. They would tend to consider many empirical based conclusions to be knowledge. Won't the AI be hindered if it treats all that as non-knowledge?

In any case The Gettier problem is refuted. When-so-ever a proposition is fully justified as true such that this justification necessitates its truth it affirms the proposition as knowledge.

You've bypassed Gettier. Gettier, I think deals with empirical conclusions. At the very least many of the Gettier problems are about empirical based conclusions. So, you have refuted Gettier's problems. You've more or less conceded the whole area he was pointing out the problems of a lot of JTBs. That JTB isnt enough to weed out mistakes.

In those cases where the justification is stochastic rather then deterministic we have reasonably plausible conjecture rather than knowledge. The boundary condition here is the knowledge of specific instances of reasonably plausible conjecture. It will know that it does not know X and will also know that X is the closest that anyone currently has to knowledge of Y.

Now we're getting somewhere. This is beginning to put it in the place human experts are. It views different types of knowledge and also different specific conclusions with varying degrees of trust, as human experts do - when they are, in fact, experts.

My model of the world as the analytic side of the analytic synthetic distinction seems to address Quine's holism. In philosophy of science, confirmation holism, also called epistemological holism, is the view that no individual statement can be confirmed or disconfirmed by an empirical test, but rather that only a set of statements (a whole theory) can be so. It is attributed to Willard Van Orman Quine.
https://en.wikipedia.org/wiki/Confirmation_holism

I can't say I fully got Confirmation holism, but the article you linked also has criticism of it. I'm not quite sure how it works in relation to your ruling out empirical conclusions as knowledge. It seems to me that confirmation holism is not taking a stand on fallibility in general, or analytic vs. synthetic. What made you bring it up? Does it relate to what we have been discussing directly?

[PeteOlcott]

My model of the world as the analytic side of the analytic synthetic distinction seems to address Quine's holism. In philosophy of science, confirmation holism, also called epistemological holism, is the view that no individual statement can be confirmed or disconfirmed by an empirical test, but rather that only a set of statements (a whole theory) can be so. It is attributed to Willard Van Orman Quine.
https://en.wikipedia.org/wiki/Confirmation_holism

I can't say I fully got Confirmation holism, but the article you linked also has criticism of it. I'm not quite sure how it works in relation to your ruling out empirical conclusions as knowledge. It seems to me that confirmation holism is not taking a stand on fallibility in general, or analytic vs. synthetic. What made you bring it up? Does it relate to what we have been discussing directly?

Because of the brain-in-a-vat thought experiment we cannot be logically certain that there are
or ever were any physically existing things. This make all "knowledge" of the world (obtained from the sense organs) less than perfectly certain.

I brought up these same ideas about the analytic versus synthetic distinction on another forum
and they were rejected on the basis that the respondent did not believe that I addressed Quine's Holism. They seemed to have Quine's Holism exactly backwards.

The five-minute hypothesis is a skeptical hypothesis put forth by the philosopher Bertrand Russell, that proposes that the universe sprang into existence five minutes ago from nothing, with human memory and all other signs of history included. It is a commonly used example of how one may maintain extreme philosophical skepticism with regard to memory and trust in evidentially derived historical chronology. https://en.wikipedia.org/wiki/Omphalos_ ... hypothesis

I remember coming up with this myself in 1993 as the logical limits of justified certainty, thus Bertrand Russell's claim of authorship seems to be an instance of this hypothesis.

Bert's version lacked a crucial detail that mine does not lack. The creation event must have been
instantaneous or there would be subconscious memories of the creation event that would leave a tell-tale sign.

[Age]

So, if what APPEARS TO BE 'physical sensations', to you, and the correspondence between WHAT APPEARS TO BE 'physical sensations' and 'this model' (whatever model that IS, EXACTLY,) of 'the world' is at best conjecture that at least SEEMS TO BE VERY RELIABLE, to you, then WHAT is the ACTUAL ISSUE here, EXACTLY?

The Gettier problem is easily abolished by defining knowledge as the subset of
justified true beliefs such that the justification guarantees the truth of the belief.

ABOLISH the absolutely STUPID notion, term, and phrase 'justified true belief', then there is NO so-called 'problem' here AT ALL.

If one just does NOT have NOR hold a BELIEF, in the beginning, then there is NOTHING to TRY TO 'justify'.

See, if one just LOOKED AT ONLY 'that' what IS ACTUALLY IRREFUTABLY True, then they would NOT be just ASSUMING nor BELIEVING some 'thing' to be true, which may in Fact NOT be AT ALL or PARTLY.

Those cases where the justification does not guarantee the truth of the belief,
yet provides some stochastic measure that the justification makes the belief very
plausible or highly likely are conjectures and not knowledge.

AGAIN, just ridding "one's" 'self' of BELIEFS here, RIDS ANY and ALL issues or problems here AS WELL.

It IS ALL VERY SIMPLE and EASY, REALLY.

[Age]

Artificial intelligence, just like the human brain, which, by the way, works EXACTLY like a computer does, can NOT DO ANY 'thing' that has NOT been PREVIOUSLY INPUTTED INTO 'it'.

Artificial intelligence can NOT do what is done at the so-called 'expert level of each field'. Artificial intelligence just COPIES what has ALREADY BEEN FOUND, DESIGNED, and/or CREATED.

Just like the human brain can and does ONLY process with 'what' has ALREADY BEEN FED INTO 'it' or is AVAILABLE TO 'it'.

That is has been refuted empirically
https://hai.stanford.edu/news/examining ... age-models
There are numerous concrete examples where a LLM solved a problem in a way that
no human ever thought of.

Talk about completely and utterly MISSING THE POINT.

I suggest READING MY WORDS AGAIN.

AND, if you EVER become INTERESTED, then let me KNOW.

Also, IF you provide examples, then we CAN LOOK AT them.

In addition to this these LLM gain many abilities that they
were not trained for. ChatGPT began to process all information much more logically
when it was taught about computer programming.

Thus, PROVING MY POINT FURTHER.

AND, IF you EVER PROVIDE EXAMPLES, then we then HAVE, AT LEAST, some 'thing' to LOOK AT, and DISCUSS.

[Age]

My model of the world as the analytic side of the analytic synthetic distinction seems to address Quine's holism. In philosophy of science, confirmation holism, also called epistemological holism, is the view that no individual statement can be confirmed or disconfirmed by an empirical test, but rather that only a set of statements (a whole theory) can be so. It is attributed to Willard Van Orman Quine.
https://en.wikipedia.org/wiki/Confirmation_holism

I can't say I fully got Confirmation holism, but the article you linked also has criticism of it. I'm not quite sure how it works in relation to your ruling out empirical conclusions as knowledge. It seems to me that confirmation holism is not taking a stand on fallibility in general, or analytic vs. synthetic. What made you bring it up? Does it relate to what we have been discussing directly?

Because of the brain-in-a-vat thought experiment we cannot be logically certain that there are
or ever were any physically existing things.

What a complete and utter STUPID and ABSURD 'thing' to SAY and CLAIM.

IF there is a 'brain-in-a-vat', THEN there OBVIOUSLY HAS TO BE physical 'things' existing, namely, 'a brain' AND 'a vat'.

It is like, with some of these so-called 'thought experiments', absolutely NO 'thought' AT ALL goes into them.

This make all "knowledge" of the world (obtained from the sense organs) less than perfectly certain.

No matter which way ANY 'thing' is 'looked' AT, and 'thought' ABOUT, there IS ONLY One 'thing' that can be PERFECTLY CERTAIN ABOUT.

The 'brain-in-a-vat thought experiment' may well HELP in leading to this IRREFUTABLE CONCLUSION, BUT that 'thought experiment' HAS TO BE 'thought' ABOUT, Correctly, FIRST.

I brought up these same ideas about the analytic versus synthetic distinction on another forum
and they were rejected on the basis that the respondent did not believe that I addressed Quine's Holism. They seemed to have Quine's Holism exactly backwards.

Just like ALL of 'you', adult human beings, in the days when this is being written, DO HAVE some 'things' COMPLETELY BACKWARDS.

The five-minute hypothesis is a skeptical hypothesis put forth by the philosopher Bertrand Russell, that proposes that the universe sprang into existence five minutes ago from nothing, with human memory and all other signs of history included. It is a commonly used example of how one may maintain extreme philosophical skepticism with regard to memory and trust in evidentially derived historical chronology. https://en.wikipedia.org/wiki/Omphalos_ ... hypothesis

As I SAID, there IS ONLY One 'thing', which can be ABSOLUTELY CERTAIN of. EVERY 'thing' ELSE can NOT be. So, can we MOVE ALONG NOW?

I remember coming up with this myself in 1993 as the logical limits of justified certainty, thus Bertrand Russell's claim of authorship seems to be an instance of this hypothesis.

Bert's version lacked a crucial detail that mine does not lack. The creation event must have been
instantaneous or there would be subconscious memories of the creation event that would leave a tell-tale sign.

Considering what the ACTUAL and IRREFUTABLE Facts ARE, EXACTLY, WHAT and WHEN the so-called 'creation event' 'HAPPENED' IS USUALLY what NONE of 'you', adult human beings, here in this forum EVER talk ABOUT or probably EVER even ENVISIONED or IMAGINED, YET.

The 'creation event' is NOT what MOST of 'you' THINK 'it' is.