Almost all internet users will come across a CAPTCHA test at some point. If you’ve ever been asked to enter a series of random letters into a textbox, you’ve seen a CAPTCHA. When a website tests your image or pattern recognition skills, you’ve just been CAPTCHA tested.
But you may not know the purpose CAPTCHA tests serve or realize just how many captcha types exist. In this article, we dive into the world of CAPTCHA tests, explain what they are, and the different CAPTCHA examples you can expect to see.
Summary: A CAPTCHA is a test that websites use to tell bots and humans apart. They’re designed to prevent bots from accessing websites. There are many different CAPTCHAs, including math problems, confident CAPTCHAs, and even simple word problems. All of them work by providing information in a non-traditional way, forcing the user to interpret what they see and deliver an accurate answer.
What is a CAPTCHA and Why Do They Exist?
CAPTCHA stands for “Completely Automated Public Turing test to tell Computers and Humans Apart.” That’s a bit of a mouthful, hence the abbreviation. CAPTCHAs exist to help websites and applications differentiate between human users and bots.
Why is that important?
It depends on the specific use of the CAPTCHA.
On the basic level, CAPTCHAs are intended to weed out fake traffic to websites. This ensures website owners get legitimate statistics to inform their marketing and development strategies. They also prevent bots from creating accounts on websites. Ticketing sites, membership programs, and polls all use CAPTCHAs to ensure only humans can sign up and make purchases.
Think of CAPTCHA tests as the first line of defense between a website and a bot.
Without them, bots would be able to access any website they want, create accounts, and make mass purchases. These issues would affect real website visitors and their experiences.
The CAPTCHA Types
There are several CAPTCHA types, each of which is designed to test the human brain while overcoming the algorithms that bots use. The effectiveness of these CAPTCHA types varies, especially as some are fairly old and can be overcome by bots using artificial intelligence and machine learning techniques.
Type No. 1 – Math Problems
One of the earliest CAPTCHA tests, math problems are exactly what they sound like. You’re presented with a simple math question, such as 4 + 3. You solve the test by entering the correct answer into a text box.
As strange as it may sound, these simple problems are often difficult for a bot to solve. The bot needs specific programming to deal with math problems. If that code doesn’t exist, the bot can’t move forward. Even so, such tests are becoming less effective as bots become more advanced.
Type No. 2 – Text-Based CAPTCHAS
Another of the earliest CAPTCHA types, text-based CAPTCHAs present a string of random letters and characters. Often, this text is obscured by an image filter or a technique that makes the letters look non-standard, though they’re still recognizable to the human eye. These techniques include varying letter sizes, rotation, and overlapping characters.
Such distortions are called “alienations.”
Interpreting letters is difficult for bots when alienation is used. Many can’t see past the distortion, even though the text string itself is fairly simple. However, humans can still interpret the characters, which they enter into a text box.
Though effect, text-based CAPTCHAs present some problems. Distorted text may be difficult for visually impaired users to read. Plus, the alienations used may make it impossible for the visually impaired to access a website using screen reading software. As such, these traditional CAPTCHAs are starting to fall out of use.
Type No. 3 – Image-Based CAPTCHAs
Also known as image recognition CAPTCHAs or Confident CAPTCHA, image-based CAPTCHAs were designed to directly replace their text-based cousins. They usually involve the test presenting a series of images in a grid. The user is given a theme and has to select the images that relate to that theme.
For example, the CAPTCHA may present nine images in a grid. If the theme is “boats”, that means you have to click on every boat in the grid. The CAPTCHA may take you through several rounds of this image recognition test until it’s satisfied that you’re able to consistently pick out images that match the theme.
Unfortunately, image-based CAPTCHAs present problems for visually impaired users. Many screen readers struggle with image interpretation, which means a CAPTCHA image could lock some human users out entirely.
Type No. 4 – reCAPTCHA
reCAPTCHA tests are interesting. They don’t appear to do much to stop bots from gaining access to a website. You’re presented with a simple box that encourages you to click it to prove you’re not a robot. You click, the reCAPTCHA verifies you’re a human user, and you’re done.
That sounds too simple, right?
Developed in 2014 by Google, reCAPTCHA is a modern CAPTCHA technique that has some clever stuff going on behind the scenes. The trick here is that the box you’re presented with also contains a checkbox.
Human users will naturally click on the checkbox when asked to click a box. Bots are more methodical, which means they’ll click the center of any box or button they’re told to click. That means a reCAPTCHA uses a bot’s instructions to do everything perfectly against it!
The good news is that you’re not necessarily locked out if you don’t click on the checkbox. Many reCAPTCHA tests follow up with a different type of CAPTCHA if you fail to click the box in the correct place.
Type No. 5 – Audio CAPTCHA
These CAPTCHAs were developed to help visually impaired users overcome the problems presented by text and image-based CAPTCHAs. They play audio clips that read out the text or letters presented in the CAPTCHA. These recordings usually feature some form of background noise, which is intended to fool bots that try to interpret the audio.
Of course, they rely on the person entering the CAPTCHA to be capable of hearing and distinguishing the audio recording from the background noise. Still, they’re fairly effective because even bots that use machine learning often struggle to properly interpret audio.
Type No. 6 – Time-Based CAPTCHA
This tests a user’s ability to fill out a form under strict time conditions.
But the test doesn’t work quite as it seems.
The timer is actually used to detect bots rather than test a human user’s speed. Bots usually fill out forms instantly, which is a giveaway. Therefore, this CAPTCHA uses the bot’s programming against it. The efficient nature of a bot results in the form getting filled quicker than a human is capable of doing.
By contrast, a human user needs a few seconds or minutes to complete a form. While you may still get locked out if you fail to meet the time limit the CAPTCHA imposes, the amount of time provided is usually more than enough to complete the form.
Type No. 7 – Invisible CAPTCHA
Invisible CAPTCHAs are unique because they seemingly don’t require the user to do anything. There are no tests involved, meaning the CAPTCHA figures out if you’re a human using other means.
These CAPTCHAs were developed by Google and have a lot going on behind the scenes. Instead of testing you directly, they analyze the way you use a website to determine if you’re a human or a bot.
For example, let’s say an invisible CAPTCHA is implemented on a job posting.
The CAPTCHA analyzes your mouse movements, typing style, and the way you use the website to figure out if you’re a legitimate user. Again, this is all about using a bot’s programming against it. Bots fill out forms and work their way through websites with heightened efficiency. Hesitation doesn’t exist and the bot will move cursors and type text faster than a human can.
As such, an invisible CAPTCHA can identify a bot because it uses an advanced version of pattern recognition.
There are several examples of CAPTCHA tests in use that we can look at.
Many websites, including Vimeo, Wix, and WordPress, use Google’s reCAPTCHA technology. They encourage you to click a box while assuming that human users will click the checkbox portion of the box.
Some apps use social CAPTCHA techniques, which offer you the ability to sign up using a social media or Google profile. Dating apps, such as Tinder and Bumble, commonly do this both as a form of bot prevention and to allow you to use your social media profiles to source images.
Several outlets, such as Ticketmaster, moved away from traditional CAPTCHA and toward a system that relies on brand recognition. Typically, these CAPTCHAs present a brand name and ask you to describe the brand in your own words. These CAPTCHAs create advertising opportunities in addition to fooling bots.
Guarding Against the Bots
CAPTCHAs exist to prevent bots from gaining access to websites and applications. The technology behind CAPTCHAs has evolved tremendously since their introduction in 2000. Today’s CAPTCHA technology often relies on behind-the-scenes pattern recognition rather than asking the user to complete a test.
CAPTCHA development will always evolve as bots become smarter. More advanced machine learning and artificial intelligence techniques may soon allow a computer program to overcome modern CAPTCHA techniques. But by that point, CAPTCHA developers may have come up with new ways to figure out if you’re a human.
As annoying as they can be, CAPTCHAs serve an important purpose. Without them, your browsing and online buying experiences would be a lot more difficult.
Frequently Asked Questions
What is the best type of CAPTCHA?
As bots become smarter, website owners have started to focus on more complex CAPTCHA types. Though there isn’t a definitive best CAPTCHA, it’s generally better to use a test that focuses on pattern or image recognition. More basic CAPTCHAs, such as word and math tests, can be solved by modern bots.
Are there drawbacks to using CAPTCHA?
There are some negatives to using a CAPTCHA test. They’re inconvenient because they create a barrier to entry. Some CAPTCHA types are also incompatible with certain browsers or with assistive devices, such as screen readers.
When was CAPTCHA developed?
The first CAPTCHA was created in 2000 at Carnegie Mellon University. Luis von Ahn, who was a member of the team behind the technology, states that they developed CAPTCHA to prevent programs from signing up for millions of accounts.
Author: Tibor Moes
Founder & Chief Editor at SoftwareLab
Tibor is a Dutch engineer and entrepreneur. He has tested security software since 2014.
This website is hosted on a Digital Ocean server via Cloudways and is built with DIVI on WordPress.
Don't take chances online. Protect yourself today:
Protect your Devices
Protect your Privacy
Or directly visit the #1: