What is artificial intelligence?

Question

I'm a little confused by artificial intelligence.

I understand this as the ability of a machine to learn new things or do different things without actually executing code (already written by someone).

In SO, I see a lot of threads about AI in games, but IMO, which is not AI Because, if it is any software, even the print command should be called AI. In games, there is only code that runs. I would call it pseudo-AI.

Am I really wrong? Should this also be considered as AI?

Answer 1

"Minsky and McCarthy, both considered the founders of AI, say that artificial intelligence is something that is done by a program or machine, if a person performed the same activity, we would say that a person had to use intelligence to perform this task . "

A more modern definition is to turn this into a head:

Artificial intelligence is something that is done by a program or machine, if a person were to perform the same action, we would say that a person does not need to use intelligence to complete a task.

Intelligence is the ability to do something that does not require reasoning. Things like understanding and generating the tongue, ordering the muscles of the legs when you walk through the floor or enjoy a symphony. You do not stop talking about it. You understand. INTUITIVE how to interpret things in your field of vision, language and any other sensory input. And you can do the right thing without reasoning. You can easily cook your entire breakfast without any reason. :-)

Doing things that require thinking or reasoning, such as playing chess or solving integrals, is something that computers can already do.

This misunderstanding of what the mind really is was costing us 60 years and a million person-years hitting our head against a wall.

Deep learning is the currently most popular expression of an alternative path to a “better kind of AI.” Artificial intuition is a special branch of deep learning, adapted to understanding the text.

The easiest way to find out if you are dealing with a classic (vain) or modern AI is whether the system requires you to supply any model of the world (MOTW). Any MOTW means that the AI ​​is limited to work in the domain specified by MOTW, and therefore is not a general intelligence. In addition, anything with MOTW is not usually intended to expand on this model; This is a very difficult task.

It’s best to start with the Mind Model (MOTM) or Training Model. They can be obtained either from neurology (complex) or from epistemology (much simpler). A well-executed MOTM can then find out everything he needs to know in order to solve problems in any domain.

The main problem for most is to find what is called a "domain-independent significance determination method." In other words, all reasonable, natural or artificial, should spend most of their time on the question "what is important."

Google is my name and “AI” for more.

Answer 2

Wikipedia says this:

Artificial intelligence (AI) is the intelligence of machines and the branch of computer science, the purpose of which is to create it.

AI textbooks define the field as “research and design of intelligent agents”, ^{[ 1 ]} where the intelligent agent is a system that perceives the environment and takes measures that maximize its chances of success.

What you are considering is more specifically called Machine Learning , which is really an AI unit. However, as can be seen from the second sentence above, "AI", considered in games, also fits perfectly into this definition.

Of course, the actual line between what is AI and what is not is pretty blurry. This is also due to the fact that everyone and his mother believe that they know what “AI” means.

I suggest you get a more scientific book (let's say classic Russel, Norvig ) to get a better idea of ​​the different fields that are present under the huge roof of what we simply call "AI".

Answer 3

Minsky and McCarthy , both considered the founders of AI, say that artificial intelligence is something that is done by a program or machine, if a person performed the same action, we would say that a person had to use intelligence to perform this task .

Answer 4

Frank and Kirt are well versed in the academic field of AI. Any difficulty in determining AI reflects the more general problem of determining real intelligence. If AI has proven anything, then we have an unlikely idea of ​​what the mind is, how amazing organisms are to solve problems, and how difficult it is to get machines to achieve the same results.

Regarding the use of the term AI in the video game industry, your turmoil is justified. The prospect of smart characters in games is so convincing that the term has long taken its own life as marketing jargon. However, AI really just badly chose a name to solve problems that computers find difficult and people find easy. And in this sense, a lot of genuine AI work is going on in the gaming industry.

Take a look at AIGameDev.com for a taste of what is currently considered worthy of attention in the development of AI games.

Answer 5

The most important aspect of AI, I think, is "curiosity." Intelligence comes from the very fact that it is the result of curiosity.

Answer 6

There is no exact definition of AI, since the mind itself is relative and difficult to determine, this is due to the fact that many areas (ancient and modern), such as philosophy and neurology, serve as the basis of AI. It depends on what is expected from your AI.

Answer 7

Artificial intelligence is an attempt to create intelligence from a computer program.

Regardless of whether it is a toy program or a neural science, as long as the program is able to imitate human problem-solving skills or even go beyond it, it is called artificial intelligence.

Of course, the expectations of computer scientists about how capable a program (or machine) can solve problems over time are increasing. Playing tic-tac-toe programs used to be considered smart until they came up with chess programs. Then we try to imitate the human brain through neural networks.

Non-professional AI is now used in most computer games. It was also used on most machines, for example, in an airplane for autopilot. A NASA researcher on Mars called curiosity (2012), which was able to detect obstacles on the ground and move around it.

Answer 8

Very complex material A.I. As if you are creating a mind that answers all the right questions with all the answers, this is AI Or just a talking encyclopedia. What if you can teach AI just by talking to him, then you think that A.I. wisely or again just a program. Perhaps a question or answer: if someone ever makes a car that looks like a man, acts a man and thinks about his man. And then see if others feel the same as his person, if they do not know him. And then if he passes this test. You see, this does not mean that the machine is conscious, or it has a mind, because they will never be answered. All this makes this a seemingly conscious act conscious, acting as if it has a mind, given that it is as close as humanity can ever understand this riddle. If a machine acts the way it cares and does useful things that have everything that really matters, not the rest of the invisible picture. We are only half to achieve this in the first place. By the way, check out Webo Learning AI Webo Learning AI