5 Answers

  1. The previous commentator is completely wrong. The fact is that artificial intelligence is a set of algorithms based on self – developing neural networks. To be more precise, artificial intelligence is a program that definitely responds to various stimuli and situations. The difference between artificial intelligence and a conventional algorithm is that artificial intelligence has the ability to self-learn based on the work of neural networks. Neural network-algorithms that recognize that if an action did not lead to the desired result, then the program will no longer do so.

    Roughly, AI is simply a set of algorithms that respond in a certain way to various situations. The same bots in computer games are not self-improving artificial intelligence.

  2. I'm not an expert, just dived a bit into this question:

    Existing achievements in the field of artificial intelligence are a number of approaches/concepts that allow you to solve complex problems – for example, automatic car control or automatic detection of diseases, playing logic games. At the moment, you can't call it “thinking” or “feeling”, it's just a very complex program that uses logical reasoning, neural networks, graphs and other mathematical tools. It is very difficult to say when what is now called AI appeared, for example, neural networks as a mathematical concept appeared 70-100 years ago, and logical reasoning was first formalized by Aristotle with his ” If a person is mortal and Socrates is a person, then Socrates is mortal.” If we talk about AI as something thinking/intelligent/self-aware, then we are far from it. It seems to me that the first problem that needs to be solved here is to learn how to extract and formalize the “meaning” of information from text/image/video/sound. For example, I would formulate the meaning of the text as something that the “writer” of this text came up with and understood/comprehended the “reader”. If you ask google translate to translate a text for you, it is precisely because of the lack of comprehension that you get a distorted result. To date, we do not understand what meaning is and how it works.

  3. Artificial intelligence is usually divided into “strong” and “weak”. A strong artificial intelligence will be a person. This is exactly what is shown in the movie Ex Machina and in the TV series Westworld. When it will be invented and whether it will be invented at all —no one knows for sure, different experts differ in their forecasts for hundreds of years. One thing is clear-a few minutes after the creation of such an AI, a technological singularity will occur.

    Weak artificial intelligence is not a substitute for human intelligence, but self-learning algorithms that solve narrow problems. This is best explained with examples. Thus, the Russian project Findface will detect human faces in pictures and finds their profiles in the social networks, the American IBM Watson detects cancer on x-rays, the application Prisma processes the user's photos in the style of famous artists, the neural network�of Yandex writes poetry in the style of Yegor Letov of�”Civil defense”, a neural network Google describes pictures Russian Abbyy develops technologies for computer understanding of natural language and semantic analysis of texts.

    Developers create a smart algorithm and “feed” it a lot of data on which it learns. Thus, computers gain the ability to independently perform intelligence-intensive and even creative tasks that were previously considered the prerogative of humans.

    The term “artificial intelligence” was first coined in 1956. AI systems of varying degrees of complexity have been developed since the mid-20th century. At the same time, the first neural networks appeared — today it is the most famous artificial intelligence technology. In general, there are several such technologies— expert systems, genetic algorithms, cluster analysis, and others.

  4. Artificial intelligence is intelligence that will be created by humans.
    It didn't appear, it will appear in 10-15 years…. over the past few years, so 50-ti.
    T. e. constantly they have 10-15 years left, but when these 15 years pass, they need another 10-15 years, and it will be so for quite a long time.
    Apparently, until scientists fully understand how natural intelligence works.
    And if by that time some genius can invent a way to somehow solve multi-purpose optimization problems…
    And if by that time some genius has figured out a way to make the computer experience at least some sensation, even the most primitive pain or the most primitive pleasure…
    And if a number of not very simple tasks are solved, then it is quite possible that AI will finally be created.
    Or maybe not.
    Maybe AI will appear only when people move from silicon computers to some quantum computers, or to some nano-bio-computers, or to some completely unknown technology.
    In general, a difficult question.

Leave a Reply