The 10-year challenge .. Have we fooled Facebook to steal more of our data?

While Facebook was looking for innovative ways to add another billion more users to its databases to take advantage of the personal information and preferences they share, the predicament was to get a new set of data without any effort. This was after the 10YearsChallange Challenge, Social networks including Facebook, Twitter, and Instagram, as well as the stories of Watts Up and Snapchat.

General challenges

The early days of 2019 bore a strange set of social networking challenges. It was the beginning with the Bird Box challenge, which came out of a movie of the same name, with the contestant licking his eyes and doing things as diverse as driving, Street, or even a tattoo on the body of a volunteer (1) (2).

Only a few hours later, another challenge on the World Wide Web was published by the world_egg_record, which published an image of an egg that, in turn, received more than 45 million impressive impressions, while more than 7 million people watched.

The strangeness of the challenges did not stop there. By the end of the second week of January of the new year, the 10-year challenge began to appear under different labels, sometimes #GlowUp and other times "#HowHardDidAgingHitYou", then spread under the "# 10YearChallenge" Or "# 10YearsChallenge". All of these labels fall under one meaning, how many have changed over the last ten years.

The user needs to enter the challenge only to put two images next to each other separated by ten years in full, and this to review the change in the form, and that's all. But such spontaneous challenges will not go unnoticed by some parties such as Facebook, so the amount of data available in the machine self-learning engineers of the machine a lot of effort.

Just another challenge?

"Memes", as you know, are never new. They are present before social networks emerge, but the new challenge has linked a wide range of recent events, including privacy and face recognition, As well as self-learning of the machine.

What users do in the 10-year challenge is to publish two personal images next to each other using special tags, which means that finding those posts by algorithms is very simple. The search spiders will archive any posts under those tags. With facial recognition techniques, algorithms can remove images that misuse these tags, such as spreading a snapshot, or any element other than human. This type of technology has long demonstrated its ability to know the human face and to distinguish between people in the picture. Available for everyone on Facebook and Google Photos, to name but a few (3).

Based on the above, what Facebook will do now is archiving images published on Instagram and its network, and then passed to the machine self-learning algorithms, which will study the details of the face of each user and the changes that occurred during that period, so you get a very large group Of data that will allow it to distinguish between people more accurately, with the possibility of knowing their age as well, a tool that Microsoft had worked on them a few years ago (4).

Conspiracy theory

The negative ideas hovering over the new challenge can be easily refuted. These images may exist primarily within Facebook, and the user is currently only collecting them together in a single picture. This is not dangerous in some people's view, but this depends on the fate of the data collected And analysis.

The studies and previous experiences in self-learning of the machine proved that the algorithms are very sensitive and quick to learn what you get. When Microsoft launched a Twitter account that responds to user tweets, it did not last more than a few hours until the company closed it completely because the algorithms learned from user feedback Insults and racist responses. In Facebook, when network engineers developed algorithms to conduct a dialogue between two computers, they developed their own language that people did not understand, prompting engineers to shut down immediately. This is a simple example of getting the machine out of text when it gets data from completely unreliable sources.

The net benefit can be targeted in the field of advertising, Facebook network is based on it, and thanks to the net data showing the change in the form of user over the past decade, the algorithms can detect some health symptoms such as dry skin for example to provide ads for products that address this. It is possible to take advantage of the geographical location of the user to link the weather factors, the degree of education, the daily activity, to display ads for sports clubs, and this is only speculation, but the Facebook network has proven over the past years the amazing ability to link user data to sell and benefit materially.

Another scenario in which data can go is, face recognition techniques are not so bad, they are currently used to detect wanted and criminals in a group of countries. As used in India, for example, to find more than 3,000 missing children. 5 When these algorithms understand changes in human form and how they grow up, they will be able to predict the shape of a person after a period of time when passing an old image, On those who are missing or who have been separated from their families by force majeure.

The good faith in the last scenario has proven its worth in Google Labs and other major companies. Artificial intelligence has defeated world champions in a set of mental games such as Go, or chess. This is after the machine watched the way they played for a long time. Analyzed them, and then developed their own methods of winning the human.

The previous scenarios do not fully disclose the intention of Facebook. But users can limit damage as much as possible. Instead of posting publicly available shares that can only be shared with friends, Instagram or Facebook will still be able to archive them, but the task of external tools and applications will be harder. Facebook learned from its mistakes and did not want to sell data to third parties without the knowledge of users.