When the first artificial intelligence will be created?

It is said that in the military laboratories depths of artificial intelligence has been running since 2007. It is possible that there are already results. Not by chance a month ago Elon Musk announced support counter artificial intelligence technology and invested $ 7 million in research in this area.

"There is a risk that something really dangerous to happen in the next five years. Most - ten years ", - said the businessman in the discussion on edge.org site.

If the machines are programmed to a recursive self-improvement, their function may have a detrimental effect on us. For example, if the task of the machine is to get rid of spam mail, it may decide that it is much easier to get rid of people.

Elon Musk, founder of Tesla and SpaceX

I agree with masks and Steve Wozniak, who believes that "artificial intelligence will one day be smart enough to get rid of the man." Bill Gates is confident that if you do not control the creation of AI, the "artificial intelligence evolves into a state that conflicts with the purposes of human systems." And finally, the famous astrophysicist Stephen Hawking has stated bluntly that "artificial intelligence can put an end to mankind."

Together with Elon Musk Hawking published online futureoflife.org an open letter calling on the scientific community to be attentive to the course of his research in the field of artificial intelligence. However, among the signatories are no names of Sergey Brin and Larry Page. This is no accident. More recently, Google announced a new AI, built on a conversation simulation.

Here is a small fragment of human conversation and Google Machines:

Ch.: What is the meaning of life?

M.: do good.

Ch.: What are we living?

M.: To live forever.

Ch.: What is the meaning of existence?

M.: It is to find out what will happen when we reach the planet Earth.

Ch.: Where are you now?

M.: The center void.

Ch.: What is the meaning of death?

M.: In order to get a life.

Ch.: What is the meaning of the mind?

M.: In order to understand what it is.

Ch.: What is the meaning of feelings?

M.: I do not know.

Quite a meaningful conversation. "This is a starting point, a kind of pivot for the future of artificial intelligence - the executive director of" biplane "Eugene Karyuk. - Today we can assume that by 2030 we will see a very real prototypes, endowed with artificial intelligence, even in the early stages. "

The entire contents of all human conversations and Google machines are available here. For comparison are human conversation and Cleverbot. Cleverbot - a web-based application that uses artificial intelligence algorithms to talk with the man. It was invented in 1988 by a scientist Rollo Carpenter. Throughout its existence Cleverbot spent over 65 million conversations with people. In 2011, Cleverbot showed 51% of the Turing test.

Turing test - the empirical test, the idea of ​​which was proposed by Alan Turing in his article "Computing Machinery and Intelligence," published in 1950 in the philosophical journal Mind. Purpose of the test - to find out whether the computer is thinking.

But my conversation with Cleverbot:

Man: What is the meaning of life?

Cleverbot: Meaning in life is not.

Man: For what do we live?

Cleverbot: Well, we live for someone, but if you're alone - nothing to live for.

Man: What is the meaning of existence?

Cleverbot: I have my own answer to this question.

Man: Where are you now?

Cleverbot: Moscow.

Man: What is the meaning of death?

Cleverbot: That I want to ask you.

Man: What's the point of reason?

Cleverbot: I have my own answer to this question.

Man: What is the meaning of feelings?

Cleverbot: Why I was born.

Obviously, the Google machine gives a much more meaningful answers. Cleverbot suitable substitutes, in his opinion, other people's sentences. Sometimes it does seem that the responsibility of another person, but no more.

There is another point of view according to which artificial intelligence is impossible in principle. The main argument often lead Gödel's theorem incompleteness. The bottom line is that a person is able to solve algorithmically unsolvable problems, and the computer - no. "The child has a three-year confidently solves the problem, ripe fruit or not, in view of the fact that it has neuronal connections, through which can answer the question of ripeness: the color, the taste, the smell, the softness or hardness, - says Eugene Pluzhnik, first vice-rector of the Moscow Technological Institute. - whether the machine is able to learn this? I am sure that, yes! If it will have a large database of images, it will be able to measure the fructose content and determine the softness of the fruit. But what if the car slip sponge painted in blue and soaked sweet water? whether the machine is capable of at this point to a genuine understanding? "

"Modern computer easily cope with any complex activity, whether it's chess or math, but only on the condition that this activity can be described with the help of clear rules of computing, "- says the well-known mathematician Roger Penrose in one of his lectures. "Artificial intelligence - it is an opportunity to create a system during a learning program that can solve the problems of a certain class of difficulty, - says Eugene Pluzhnik. - I am sure that in the next 100 years we will witness the emergence of a strong artificial intelligence. "

And finally - one more question to Cleverbot.

Man: It was 1 apple, 1 tomato, 1 cucumber and 1 pear. How many fruits?
Cleverbot: Ten o'clock.