If AI is so clever, why can't it solve a CAPTCHA?
CAPTCHAs (Completely Automated Public Turing test to tell Computers and Humans Apart) are special puzzles on the internet made to tell the difference between humans and computers. They’re meant to be tricky for machines but easy for people to solve.
Here’s why AI, which is like a very clever computer, sometimes has trouble with CAPTCHAs:
HARD TO SEE: CAPTCHAs often show wobbly letters, numbers or pictures that are hard to read. While AI is good at recognising normal images and text, CAPTCHAs are made to be confusing so that computers can’t easily figure them out.
ALWAYS CHANGING: As computers get better at solving certain CAPTCHAs, new and harder ones are created to stay ahead. This means that CAPTCHAs are always being made trickier to keep computers from solving them.
HOW AI THINKS: AI doesn’t ‘see’ things like humans do. It looks for patterns in what it’s been taught. If a CAPTCHA breaks these patterns by being extra confusing, the AI might not be able to solve it.
KEEPING THINGS SAFE: The main job of CAPTCHAs is to stop robots (automated programs) from doing things like making fake accounts or sending lots of spam. Even though AI can sometimes solve CAPTCHAs, letting it do that would ruin the whole point of using them for security.
So, even though AI is very clever, CAPTCHAs are designed to be tricky on purpose, so computers can’t easily get past them. That said, AI is getting better and better at this. In fact, a 2023 study showed that most AI can solve CAPTCHA with a 96% accuracy – while humans’ accuracy ranges from 50-86%!
