I saw a pretty funny tweet over the weekend about what it means to be human.
Basically the joke was about the Turing test. If you don’t know, the Turing test was developed by Alan Turing in 1950 as a way to prove whether or not we had in-fact achieved a legitimate version of artificial intelligence.
The basic idea is that if a machine and a human interact with another human and that human can’t tell the difference between the two, then the Turning test is passed.
The joke is that we used to think the Turing test was proper validation criteria for what it meant to be human, but it’s actually about whether or not we can identify pictures of a boat.
They were referring to the concept of the CAPTCHA test we see across the web. You know where you’re shown 9 pictures in a 3 x 3 grid with and without bicycles in them and you have to pick all the ones with a bicycle?
Nerdy humor I know, but I promise there’s a point to all this.
Get our awesome product content delivered daily-ish to your inbox
Turns out, this can actually be beaten rather easily by neural networks and deep learning. I guess we need a new test?
To learn more about how neural networks can beat the CAPTCHA, watch the video linked below.