AI 2023. Meet ChatGPT. - page 206

New comment
 
Aleksey Nikolayev speed of change since the 19th century is quite impressive, especially compared to the speed of biological evolution.
From what I understand, the main difficulties of powder metallurgy are the cost and complexity of the technical production of metal powders. Plus, the limitation on the size of the parts. Mass production turns out to be unprofitable.

3D printing essentially doesn't change anything. Just a more modern version of smelting.
 
By the way, 3D machining of metal parts has long been realised on CNC machines. There is no AI there).
 
Реter Konow #:


With all the "power" of modern AI, it cannot solve any task facing specialists and is useless for front end developments. To suggest/draw something obvious and average - please, but to find a new solution - sorry). To solve high school students' problems - sure, to make a discovery in the scientific field - no way.

I think we are approaching the point where the computing power of computers can no longer solve anything. I'll explain in more detail why.

Type in any search engine something like - the use of AI in science, the results will probably surprise you. For example, here is the first thing that came up in physics.

-----------------

AI in physics has long been used to analyse big data. And it has a lot to boast about. In 2012, machine learning models helped staff at the European nuclear research centre CERN discover the Higgs boson. The task of AI was to analyse the endless stream of signals from the Large Hadron Collider, look for signs of this elementary particle and mark them.

In the long term, AI can simplify the solution of quantum problems. Proof of this is the work of researchers from New York: they created and trained an algorithm that reduced the calculations of the Hubbard model from 100,000 equations to four. The accuracy of the calculations did not suffer.

Another possible future task for AI is to find new physical laws. To make this a reality, we need an algorithm that can detect state variables. And scientists at Columbia University have succeeded. Their AI was able to independently guess what sets a pendulum and a lava lamp in motion, as well as why a fireplace is burning. Of the inputs, the tool had only video recordings. The variables suggested by the artificial intelligence did not always coincide with those to which the physicists themselves are accustomed. The scientists concluded that AI has a chance to show people previously unknown driving forces of nature and to push new conclusions that will probably change both science and our view of the world.

 
Реter Konow #:
Mass production is unprofitable.
Profitable.
A lot of sanitary products are made of powder.
 
Speaking of nanotechnology. At one time, I remember in the noughties, if I'm not mistaken, they were the Grail. Like AI is now. Everyone was talking about them. Then they just kind of forgot about it. Nowadays you hardly ever hear about new breakthroughs in nanotechnology. The market picked up on it and hyped it up. Many made good money on the ignorance of the masses.

Of course, the topic of nanotechnology is still alive, but like everything else, it depends on the profitability of mass production.
 
Sergey Gridnev #:
Cost-effective.
A lot of plumbing products are made from powder.
I was referring to the unprofitability of parts with complex properties from different alloys. Simple metal powders have been produced since antiquity.
 
sibirqk #:

Type in any search engine something like - the use of AI in science, the results will probably surprise you. Here is for example the first thing that came up in physics.

-----------------

AI in physics has long been used to analyse big data. And it has a lot to boast about. In 2012, machine learning models helped staff at the European Centre for Nuclear Research CERN to discover the Higgs boson. The task of AI was to analyse the endless stream of signals from the Large Hadron Collider, look for signs of this elementary particle and mark them.

In the long term, AI can simplify the solution of quantum problems. Proof of this is the work of researchers from New York: they created and trained an algorithm that reduced the calculations of the Hubbard model from 100,000 equations to four. The accuracy of the calculations did not suffer.

Another possible future task for AI is to find new physical laws. To make this a reality, we need an algorithm that can detect state variables. And scientists at Columbia University have succeeded. Their AI was able to independently guess what sets a pendulum and a lava lamp in motion, as well as why a fireplace is burning. Of the inputs, the tool had only video recordings. The variables suggested by the artificial intelligence did not always coincide with those to which the physicists themselves are accustomed. The scientists concluded that AI has a chance to show people previously unknown driving forces of nature and to push new conclusions that will probably change both science and our view of the world.

Certainly for science - AI is a good tool.

A tool, .... not a scientist.))
 

Реter Konow #:

3D printing doesn't really change anything. Just a more modern version of smelting.

It does. Obviously, it gives (in the distant future, of course) a fundamental possibility of self-reproduction of robots without the participation of humans. Powder is poured in, the necessary programme-model is loaded, and a new robot comes out).

Of course, we still need nanotechnology of sufficient level to print electronic brains, but if you look closely at the current technology of chip production, it is obvious that much has already been invented.

 
Aleksey Nikolayev #:

Changes. Obviously, this gives (in the distant future, of course) a fundamental possibility of self-reproduction of robots without human participation. Powder is poured in, the necessary programme-model is loaded, and a new robot comes out).

Of course, we still need nanotechnology of sufficient level to print electronic brains, but if you look carefully at the current technology of chip production, it is obvious that much has already been invented.

But powders have to be produced. And who will do it? Robots? And who will carry out nanotechnological processes in sterile laboratory conditions? )))

The point is that the magic 3D printer printing all the parts of the robot, from the body to the processors, will be so complex that mass production will quickly "kill" it. Although it may be enough for a couple or three robots.)))
 
Реter Konow #:
Certainly for science - AI is a good tool.

A tool,.... not a scientist.))

You didn't read the last paragraph carefully:

"Scientists have concluded that AI has a chance to show people the previously unknown driving forces of nature and push for new conclusions that are likely to change both science and our view of the world."

And this is already now, and what will happen in three years, when heterogeneous AI systems, for example, will be united into a single complex, which will be able to find new patterns, investigate them and then report the results.

And in six years, or nine years?

1...199200201202203204205206207208209210211212213214
New comment
Reason: