Your call if to use; newest, current or yesteryear tech.
When i need something, i look what is currently available and get that. Rather than wait for better deals. Since one can remain waiting for forever.
that is a pragmatic approach, and is advisable eg for investment, eg once there were 7% investments available, I wanted 8%, waited for this, but they gradually dropped over a few years to 5.9%!
This reminds me the fiasco Amazon had with their "AI powered" instant check-out system.
Article: https://arstechnica.com/gadgets/202...e-checkout-which-needed-1000-video-reviewers/
Whereby:
As far as AI goes, it's another step in Technological Singularity,
wiki: https://en.wikipedia.org/wiki/Technological_singularity
1st we had PCs, that made several calculations much faster than humans, thus, dedicated people needed to calculate stuff, lost their job.
2nd step was robotics (automation), where line workers in factories were made essentially obsolete.
Now, 3rd step is AI, that makes the thinking for humans.
E.g AI art that won an art prize:
Article: https://www.nytimes.com/2022/09/02/technology/ai-artificial-intelligence-artists.html
And of course, AI usage has become cheaper, faster and better quality than that of a humans. E.g article where game developer payed AI artist for the artwork, rather than hiring artists to do the drawings,
article: https://www.pcgamer.com/games/card-games/champions-tcg-ai-artist/
ok, AI isnt just AI, its an umbrella term for a number of incomparable technologies!
each of which has to be discussed as a separate discussion.
that art is generative AI, but for me that isnt true intelligence, but is just parametrisation,
plus voice recognition for voice based generative AI.
the big problem is pattern based AI, as patterns NEVER determine truth, they only hint at truth.
I did pure maths at uni, and the first term of the first year we were warned about the danger of patterns,
no amount of patterns makes something true, in pure maths you have to use logic, not patterns.
a famous example from history is that n^2 + n + 41 is a prime number for
n=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39
so the pattern is it is always prime,
the problem is it isnt prime for n=40, because 40^2 + 40 +41 = 40x(40+1) + 41 = 40x41 + 41 = 41x(40+1)=41x41
in pure maths in an exam if you argue by pattern you will fail!
because it is established beyond all doubt that patterns are not truth.
the problem with Chatbots is they are truth by democracy, but truth isnt a democracy.
most opinions on the internet are wrong. you have to trawl through a lot of junk to reach the gems.
wisdom also isnt a democracy, most people make one bad decision after another, including myself!
in 1800, all scientists thought that time was absolute, so if you had AI in those days, the chatbot would tell matter of fact that time is absolute. but then along comes Einstein, with his theory of special relativity, which says that time is relative to the observer, it isnt absolute.
you can travel 1000 years into the future in 1 second, if you have enough money and technology, basically as you approach the speed of light, time slows down to 0 for you.
this is a big problem that pattern based AI is making all the decisions now, and in particular AI doesnt understand the role of an employee, because there isnt a direct link between an employee and money.
eg the guy stacking goods in a supermarket, no money flows via that guy.
the AI then decides to sack everyone, as then much more money to be made.
calculators and robots aren comparable, as calculators and ordinary software are based on logic, so remain in truth. robots generally just are fed in the motions of humans for fixed tasks,
the problem is when you bring in patterns.
eg the british navy long ago decided to use neural net AI for their warships, they trained the neural nets, then put them to use, and the AI began shooting down their own fighter planes!
they troubleshot this, and found it was because the training data was all daytime images, the planes shot down were at night, the nightness was regarded as not fitting the pattern and the AI shot down the planes!
this isnt a one off, but AI ongoingly makes such blunders, eg one New York lawyer used a chatbot for his case, and got into professional trouble because the AI was quoting cases that never existed! it basically extrapolated from cases that did exist! the AI lacks the common sense that you cannot just invent case law! case law is entirely about real cases that went to the appeals courts.
another example is an australian mayor is suing Chatbot for defamation, as it alleged he had done things which he hadnt done. the thing is a chatbot will blindly believe everything on the internet, including disinformation.
the chatbots trawl the news media. but all journalists as part of their training always insert fake facts into news stories, this is to prevent people acting on the info. this can be anything from a person's name, or age, or interests or quotes etc.
so the chatbots will fill up with false info.
anyway, its not just a progression from calculators to robots to AI,
the pattern based AI is a provably false MO, it is guaranteed to fail, and already has in specific cases.
see the above eg of n^2 + n + 41 as to why patterns are a very unreliable MO. and this example was discovered in a bygone era as a warning as to the danger of relying on patterns.
each of which has to be discussed as a separate discussion.
that art is generative AI, but for me that isnt true intelligence, but is just parametrisation,
plus voice recognition for voice based generative AI.
the big problem is pattern based AI, as patterns NEVER determine truth, they only hint at truth.
I did pure maths at uni, and the first term of the first year we were warned about the danger of patterns,
no amount of patterns makes something true, in pure maths you have to use logic, not patterns.
a famous example from history is that n^2 + n + 41 is a prime number for
n=0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39
so the pattern is it is always prime,
the problem is it isnt prime for n=40, because 40^2 + 40 +41 = 40x(40+1) + 41 = 40x41 + 41 = 41x(40+1)=41x41
in pure maths in an exam if you argue by pattern you will fail!
because it is established beyond all doubt that patterns are not truth.
the problem with Chatbots is they are truth by democracy, but truth isnt a democracy.
most opinions on the internet are wrong. you have to trawl through a lot of junk to reach the gems.
wisdom also isnt a democracy, most people make one bad decision after another, including myself!
in 1800, all scientists thought that time was absolute, so if you had AI in those days, the chatbot would tell matter of fact that time is absolute. but then along comes Einstein, with his theory of special relativity, which says that time is relative to the observer, it isnt absolute.
you can travel 1000 years into the future in 1 second, if you have enough money and technology, basically as you approach the speed of light, time slows down to 0 for you.
this is a big problem that pattern based AI is making all the decisions now, and in particular AI doesnt understand the role of an employee, because there isnt a direct link between an employee and money.
eg the guy stacking goods in a supermarket, no money flows via that guy.
the AI then decides to sack everyone, as then much more money to be made.
calculators and robots aren comparable, as calculators and ordinary software are based on logic, so remain in truth. robots generally just are fed in the motions of humans for fixed tasks,
the problem is when you bring in patterns.
eg the british navy long ago decided to use neural net AI for their warships, they trained the neural nets, then put them to use, and the AI began shooting down their own fighter planes!
they troubleshot this, and found it was because the training data was all daytime images, the planes shot down were at night, the nightness was regarded as not fitting the pattern and the AI shot down the planes!
this isnt a one off, but AI ongoingly makes such blunders, eg one New York lawyer used a chatbot for his case, and got into professional trouble because the AI was quoting cases that never existed! it basically extrapolated from cases that did exist! the AI lacks the common sense that you cannot just invent case law! case law is entirely about real cases that went to the appeals courts.
another example is an australian mayor is suing Chatbot for defamation, as it alleged he had done things which he hadnt done. the thing is a chatbot will blindly believe everything on the internet, including disinformation.
the chatbots trawl the news media. but all journalists as part of their training always insert fake facts into news stories, this is to prevent people acting on the info. this can be anything from a person's name, or age, or interests or quotes etc.
so the chatbots will fill up with false info.
anyway, its not just a progression from calculators to robots to AI,
the pattern based AI is a provably false MO, it is guaranteed to fail, and already has in specific cases.
see the above eg of n^2 + n + 41 as to why patterns are a very unreliable MO. and this example was discovered in a bygone era as a warning as to the danger of relying on patterns.