Every time you tick a box that says “I’m not a robot” or select all the pictures with traffic lights, you’re taking a tiny test called CAPTCHA — short for Completely Automated Public Turing test to tell Computers and Humans Apart. Its job is simple: to check whether you’re a real person or an automated bot.
Bots are computer programs that can perform repetitive tasks at lightning speed — like creating fake accounts, posting spam, or stealing information. Websites use CAPTCHA tests to stop them from pretending to be human users.
The early versions of CAPTCHA showed wobbly letters and numbers that humans could read but bots couldn’t. But as AI improved, machines started learning to recognise distorted text too. So, newer versions rely on image-based puzzles — like spotting bicycles or crosswalks — because humans are still better at understanding context than most bots.
Some CAPTCHAs, like the “I’m not a robot” checkbox, don’t test what you click — they analyse how you move the mouse. Real users move naturally, while bots follow perfectly straight, mechanical paths.
Invisible CAPTCHAs go even further. They work quietly in the background, checking your browsing behaviour — like how fast you scroll or how long you stay on a page — before letting you in.
So the next time a CAPTCHA pops up, remember — it’s not just testing your patience. It’s protecting websites, keeping bots out, and giving humans the chance to prove they still browse the web differently.
There’s an invisible version
Some CAPTCHAs work silently, analysing your typing speed or mouse movement without showing any puzzles.
Puzzles have saved books
In early reCAPTCHA versions, users helped digitise old texts by identifying faded or unclear words from scanned books.