How does Captcha work?

NoCAPTCHA, reCAPTCHA

While using the internet, logging in to websites, purchasing tickets you've probably seen a CAPTCHA! But what is it and why is it everywhere?

CAPTCHA stands for "Completely Automated Public Turing test to tell Computers and Humans Apart."

CAPTCHA is basically a test that was introduced by Luis Von Ahn and his team of researchers. The point of making CAPTCHA was to stop spam bots or automated bots. This also prevented DDOS attacks on websites. Like if you look at a ticket booking site (where automatic ticket buying bots are a huge problem) you will see a CAPTCHA, this is to prevent bots from buying out all the tickets and selling them at a higher price.

So due to this, many sites started using them and soon enough millions of people on the internet started solving puzzles that computers won't have been able to solve easily. So Luis Von Ahn thought that how can they use this to their advantage and do something more productive.

So they started utilizing this power by digitizing books and other kinds of documents, and so we started seeing parts of books that were scanned using OCR (Optical Character Recognition Software) and were too hard for the computer to solve it.

Picture From "Zuck That" Youtube Video

Luis called this project "reCAPTCHA."

Due to the huge number of people solving CAPTCHA every single day, Luis and his team were able to digitize around 2.5 million books everyday! This is an insane number and shows how powerfully can one use the internet.

Later in 2009 - Google saw the potential of reCAPTCHA and acquired it. Google used this brainpower for digitizing several different documents and, then when they ran out of documents to digitize, they started digitizing street signs, which helped Google Maps.

But clearly there's a big problem with this! Though CAPTCHA might seem easy for a normal person, it was very difficult for the disabled (mainly people who were blind or had dyslexia) to solve it! And so there were audio versions of CAPTCHA. These were CAPTCHAS in the form of audio, but the audio wasn't very clear and straight forward - The audio was weird enough that a computer wouldn't be able to understand but on the other hand it was not so weird enough that an average human with hearing capability could understand it!

An another major problem with CAPTCHA was that as time was moving on, bots were getting smarter and so there image recognition and "text reading" skills were improving as well, this meant that the CAPTCHA's was being much less effective and bots were able to bypass CAPTCHA which was a massive problem.

In order to solve this problem, engineers started making CAPTCHA more harder to read, even for some humans, by adding weird effects to the letters, by adding noise and different kinds of backgrounds. That did solve the problem - but soon bots were able to solve those as well - In fact at this point Computers were able to read these better than humans could!

In order to solve this problem Google came up with "No CAPTCHA reCAPTCHA."

And these are the ones that we see today. But how do these work then. Any one can just click on a box and wait for a second or two, even this super cool swag bot can!

Well Yes, - But there is a lot going on behind the scenes. Just moments before you click the box your activity is tracked. It is checked if your activity seems natural or "bot-like."

Things like how you scroll, bring your cursor to the box and many more variables (Google doesn't reveal all of them) are checked and if it seems more human like then you'll be able to pass! Sometimes Google's servers aren't sure if you're a robot or not, so you get an image CAPTCHA, like select the pictures with a shop in it.

And this method actually is better and faster than the methods before. This solved problem for people with disabilities and it reduced the time it took read those annoying CAPTCHAS.

And this is how CAPTCHAS are now!