Since this Wednesday, video platform YouTube has been using Artificial Intelligence (AI) in the United States to check the age of users. “We are now bringing it to the US and as we make progress, we will roll it out in other markets,” thus James Berer, product manager at YouTube Youth in an earlier explanation.
With the AI control, Youtube wants to prevent minors from watching sexually tinted images or videos with social aggression, reports YouTube official blog.
The tool checks which type of videos is being watched since when the account exists and what the search history is. So if you say you are eighteen, but only to the cartoon Peppa Big or game videos over Minecraft Looks, the AI tool can ring the bell. He then asks the user about a photo of your proof of identity, credit card details or a selfie via the webcam.
Since the announcement of this verification at the end of July, many minors have wondered online how they can circumvent the new gatekeeper: they are talking about it on the online social platform Reddit, among other things. “The camera control detected that I am eighteen years or older,” claims a minor Dutch teenager in the largest Youtube-Fangroep On Reddit.
1
The new AI tool will find out the age of users with self-learning algorithms. How does that work?
The AI tool is a model that estimates and uses ages Machine Learning Systems To interpret different signals. These systems ‘learn’ themselves to which formulas they should apply, such as the entered age in the account and the YouTube activities. “Our age assessment model does not collect new information that has not yet been linked to the account,” says a spokesperson for Google Benelux, the owner of YouTube.
2
What if a YouTube account from an adult is accidentally mistaken for a child-because it plays a lot of videos for children?
The video platform is ‘convinced’ that the signals that the new AI tool will also receive the cases of co-watching (look together) Correctly recognize. “If our model makes the wrong estimate in a certain case, you can always have your age verified with an official identification, a selfie or your credit card details,” says the Google spokesperson. In Europe Also checks YouTube the data from users, but without the help of AI; When in doubt, their private data is also asked.
3
Why is Youtube introducing the extra control now?
The new gatekeeper was set after several countries entered stricter legislation on online age control. Like this announced To ban Australia to YouTube at the end of July for teenagers under the age of sixteen. And in the United Kingdom the Online Safety Act In force that must ensure that minors do not see any pornographic material.
But minors know how to circumvent the measures through, for example, VPN services (Virtual Private Network) to use their location. This makes it look like they use the internet from another country. Or they show images of adult faces during a camera control. According to Liem, control will always remain a “cat and mouse game”. She compares it with adblockers: an application that is used by users to bypass advertisements.
4
Why does Youtube choose to moderate the age control itself, and not to wait for regulations by governments?
The parent company Alphabet tries to keep in control as much as possible. The AI tool is an attempt not to lose this check, Liem suspects. But the question is whether the AI tool videos that are uploaded under the guise of ‘child-friendly’-in which violence actually occurs-can detect. Liem: “The new AI tool could only be introduced in Europe if it is in line with the European AI Regulation [een Europese wet die regels stelt voor de ontwikkeling en het gebruik van AI-systemen]. It is not yet possible to say whether that is the case, because it has just been entered for one day. “