Table of contents:
Facebook is migrating to a much more private and user-centric platform although a new report from Reuters says they are viewing private posts to train their Artificial IntelligenceLike many other Internet companies, Facebook uses AI for certain tasks that help it categorize content.
Training these artificial intelligence algorithms is not a simple task and humans are helping in this process.The source talks about several companies that work for Facebook Their mission is to tag posts by hand to check the effectiveness of the algorithms and improve them.
Facebook uses Artificial Intelligence to protect users
Among other tasks, Facebook also makes use of Artificial Intelligence to protect users and remove sensitive content Its algorithms not only work on Facebook but they can also collaborate on Instagram and even WhatsApp, an application where sexual content continues to be the order of the day.
The "problem" with these trainings is that Facebook is reviewing users' private content and that's not good for anyone. The platform should only use its algorithms with public content, especially if these posts are being monitored by humans.Facebook ensures that this work is completely legal and defends this type of practice. Despite this, they could be breaching the harsh GDPR regulations of the European Union, much stricter than that of other regions of the planet.
Artificial Intelligence is the present
When you use a CAPTCHA in which you help to identify contents within an image you are improving the algorithms in their task of identifying certain objects . All this type of work is necessary to be able to perfect the algorithms and that some devices such as security cameras can perfectly distinguish between a dog and a human (for example). AI has come a long way in recent years and must continue to follow suit.
Tagging posts with humans is being very productive when training these systems. Facebook uses it for tasks as necessary as detection of adult content on its platform.
