Facebook, a social network platform that belongs to Meta, asks users to upload images from their phones to offer collages, resumes and other ideas using artificial intelligence (AI), including those who have not been downloaded directly to the service.
According to Techcrunch which First reported A feature that users submit a new pop -up message asking you to “allow cloud cleaning” when they try to create a new Facebook story.
“To create ideas for you, we choose the media from your camera and load it into our cloud constantly, based on information, such as time, location or topic,” the company said. “Only you can see the suggestions. Your media will not be used for targeting. We check this for safety and conscientiousness.”
If users agree to make their photos processed on the cloud, Meta also states that they agree to their AI CONDITIONSWhich allow you to analyze your media and personality traits.
On the help page, meta speak “This feature is not yet available to everyone,” and that it is limited to users in the US and Canada. He also noted that Techcrunch, that these AI offers are disconnected and can be disabled at any time.
Development is another example of how companies seek to integrate AI functions into their products, often through users’ privacy.
Meta says its new AI feature will not be used for target ads, but experts are still causing problems. When people load personal photos or videos – even if they agree with it, it is unclear how long this data is stored or who sees them. Because processing occurs in the cloud, there are risks, especially with things such as personality recognition and hidden details such as time or location.
Even if it is not used for advertising, such data can still be in training sets or used to create user profiles. It’s a bit like the transfer of your photo algorithm that quietly studies your habits, advantages and samples over time.
Last month is meta undertake To prepare their AI models using public data that are divided by adults on their platforms in the European Union Received approval from the Irish Data Protection Commission (DPC). Campaign stopped Use Generative tools II in Brazil in July 2024. In response to Privacy Problems grown by the government.
Giant Social Media also added AI functions to WhatsApp, the latter is the ability summarize unread messages In chat using a privacy approach, he calls private treatment.
This change is part of a greater trend in the generative II, where technology companies mix convenience with tracking. Functions such as automatically made collages or reasonable story suggestions may seem useful, but they count on AI, which looks like you use your devices-not only app. That is why privacy settings, obvious consent and restriction of data collection are more important than if it is.
AI Facebook feature also comes as one of the German Data Protection Watchtime designed to remove Deepseek apps from the relevant applications stores from the unlawful data transfer to China following similar problem grown several countries At the beginning of the year.
“Service of extensive personal data users, including all text records, chat history and downloaded files, as well as location, devices used and networks,” according to A, according to A statement Published by the Berlin Authorized Data Protection and Freedom of Information. “The service transmits the collected personal data of users to the Chinese processor and stores it on servers in China.”
These translations violate overall data protection regulation (GDPR) The European Union, given the lack of guarantees that German users’ data in China is protected at an equivalent block.
Previously this week Reuters report that the Chinese company AI assists the country’s military and reconnaissance operations, and that it exchanges user information with Beijing, citing an anonymous Department of the US Department of State.
A couple of weeks ago also Openai land In $ 200 million with the US Department of Defense (DOD), “to develop prototypical Border Guards AI to solve critical issues of national security both in hostilities and in enterprises.”
Campaign – Note This will help the Pentagon “to determine and prototype as a Frontier AI can convert their administrative operations: from improving how members of the service and their families receive medical care, to ordering how they look at programs and acquisition data, to support the active cyber -defense.”
