Captcha has become an integral part of our lives. If you have not come across it, you are either lucky, or you simply do not exist. Small pictures on which it is necessary to note cars, bikes, traffic lights and so on, occur always and everywhere. Especially often this sin Google. I think they need to understand not a robot do you? Partly Yes, but they also have another secret purpose. Put on a tinfoil hat and read on.
May says it looks a little odd and far-fetched, but we have specific facts that make it clear exactly where “Corporation of Good” use we convert our spare time in their own benefit. Personally, I’m not nice and I didn’t want the events to continue developing, but so far all is well, and may soon become worse. Okay, first things first…
What is captcha
For a better understanding of everything discussed below, consider first a definition of what we used to call such a word as “captcha”.
The word is not Russian and in the original, is an abbreviation of several words in English. It is written as CAPTCHA, and the deciphering of sounds like “Completely Automated Public Turing test to tell Computers and Humans Apart” (completely automated public Turing test to distinguish computers and humans).
The Turing test is an empirical test, the idea of which was proposed by Alan Turing in the article “Computing machinery and intelligence” published in 1950 in the philosophical journal Mind. Its purpose was to determine the ability of the machine to enter the man astray into thinking that it is communicating with another person.
This principle lies at the basis of the algorithm that determines who tries to send a request to receive information from the server. As a rule, there are three main types of captcha. It’s badly written of ciria or letters which would not be able to determine the computer that the mathematical operations that the computer is easy to perform, but do not realize that they have to perform and the definition of objects, as a favorite Google.
How does captcha
We will focus on the third option with pictures. In order to obtain access to information, we need to find the pictures of certain objects. For example, bicycles, traffic lights, shop Windows and so on.
Normal procedure, you say, and partly you will be right. But have you ever wondered why when a partially incorrect answer, you are still allowed to some information? It turns out, no matter whether we say? This can be explained by the fact that the algorithm lies a certain margin of error that is allowed when taking this test. Suppose this is true, but why this way often protected the simple results of a search engine? Nothing will happen if the conventional robot learns how many stars in the constellation Orion? Protection from DDoS-attacks? Perhaps, but there is a more simple explanation.
What is AI
Google, like other industry giants, is working to create his version of artificial intelligence, or as it now often called, AI. I’m a little skeptical of this phrase, but let’s leave such arguments for another article.
In the modern sense of the term AI it is the ability of the computer to simulate human activity, including, through machine learning to learn to identify objects.
Usually, for such systems, they must first teach by example. That is, the person shows the system bus and says it’s a bus, then shows the fire and says it’s a fire. This continues for some time, then the system tries to determine where what a person says is right or not.
It is possible that the captcha system from Google aimed at precisely that. Considering the billions of queries that people around the world send to the server, the training can be very good and complete. Even if you offer a captcha every tenth user will still be enormous amounts of data that the company can use to train their systems. Thus, in fully automatic mode.
As Google wants us to use
Of course, it’s not officially confirmed, but judging by all facts the company is using us. Confirmation is that the captcha appears so often and for some reason sometimes ignores errors.
If this is true, just imagine how much money the company saves on the army of experts who must train the system, recognize image. The same refers to cases when you are invited to read the blurry text on the facades of houses. In this situation, the system can learn to identify numbers while scanning the streets for various map services.
For example, in the picture above are some examples of captchas with house numbers. On the one hand it’s just not a very contrasting image that most likely will not be able to recognize the robot. On the other hand, why the numbers? House numbers — to this maps from Google. The creators of the service will not have to place the house separately. The system itself can understand where some number and plot it in automatic mode. But we then also check the result when you find the right home and send edits for the consideration received as a reward ephemeral status as a member of team Google. Because house numbers are not painted, but photographed really and most likely the cars Google, which all drive on the roads, removing street for “panoramas”.
In the future it will help the work of Autonomous cars. To confirm this, it is worth remembering the picture of the captcha with traffic lights, pedestrian crossings and intersections. Such pictures not in the best quality will issue a camera and autopilot system will have to recognize them. While stand-alone machines in a broad sense, but the data is already piling up.
Additional evidence that this is exactly what Google is trying to pump up their services information are the Chinese system of identification, which are reduced to simple operations, like checking in the right place or swipe the screen. They do not work for Google and they don’t have to engage in such data collection.
In addition to images, Google also offers recognition of the voice text. You know what I mean? Google is your assistant, which for teaching need someone to communicate and not only listen to requests, but also to understand how well the person understands what he said. For this and uses a system voice captcha.
If this is true that Google makes us work for them? It turns out! Just imagine what contribution each of us makes a big deal of teaching the AI from Google, while free. I am personally a bit annoying after each request to Google to decipher the pictures, but I noticed that it goes from one user to another. One time no one understood me when I said that almost every search query from a smartphone for me is accompanied by a captcha.