Google plans to make CAPTCHA frustrations a thing of the past with its new system, Invisible reCAPTCHA. Rather than making people decipher a piece of text or other media, as is the norm, Invisible reCAPTCHA will judge whether or not you are human without having to interact with you at all.
Algorithms in Google’s Advanced Risk Analysis technology will make the judgement based on a user's behaviour. It’s a step up from Google's previous No CAPTCHA ReCAPTCHA, which asked people to tick a box saying “I’m not a robot”, but uses the same engine to process the information and protect services from unwanted spam.
It’s unclear how Invisible ReCAPTCHA will work, and how much data it needs to scrape on a site visitor in order to verify their warm-bloodedness. When Google introduced the ReCAPTCHA tick-box, and with it, its Advanced Risk Analysis system, it described the new reCAPTCHA as: "actively consider[ing] a user’s entire engagement with the CAPTCHA — before, during, and after — to determine whether that user is a human". Since Google currently says “reCAPTCHA is the most widely used CAPTCHA provider in the world”, that’s a lot more user data the search giant will be able to gobble up.
The service is free, and has always appeared to contribute a lot of good: filtering out bots while simultaneously training the future of artificial intelligence and even saving books; when users solve a word CAPTCHA, it may be coming from a book where the AI cannot read the text but is working on digitising it for audiences.
WIRED has reached out to Google in order to find out how Invisible ReCAPTCHA will work, how much user data it will vacuum up, and what it plans to do with that data. For now, if you want to integrate the new reCAPTCHA to your site or app you can register here.
This article was originally published by WIRED UK
