European users of the social networks Facebook and Instagram have received a notification in recent days informing them that from June 26, their data may be used “to develop and improve Meta’s artificial intelligence (AI)”the parent company of these two platforms.
Thursday June 6, the Viennese association None of your business (NOYB), specialist in the protection of personal data, filed a complaint with the National Commission for Information Technology and Freedoms, as well as its equivalents in ten other European countries . She denounces a use “illegal” of this data and asks the privacy police to intervene ” emergency ” to prevent the application of this new regulation.
What do these changes actually mean for European Meta users? Explanations.
Why is Meta interested in our data for AI?
In a press release published on May 22, Mark Zuckerberg’s company explained its intention to offer, by the end of the year in Europe, “new generative AI experiences” within its applications, already available in other geographic areas. But for that, she says, these AI-based products “must be trained on information that reflects the diversity of cultures and languages of the European communities who will use it”.
To function, AI programs must be trained on enormous masses of data, generated by humans. The extensive information that Facebook and Instagram users publish on these platforms is a gold mine for Meta, a leader in the field of AI.
However, the European Union is more protective than others on the issue of its citizens’ personal data. This explains why Meta is implementing this new policy later than in other regions of the world, such as the United States, and allowing European users to refuse the exploitation of their data to train its AI – a choice that no The Americans didn’t get it.
What data will Meta use to train its AI?
The company will notably use “information shared by Internet users on Meta products and services”, such as texts, photos, comments or even stories (temporary publications) published on Facebook or Instagram. However, she promises that she will not use “the content of your private messages”. The company will also operate “the information people share when they interact with our generative AI products”like conversational robots for example.
The company adds that it may also collect data that does not belong to users of its services: “For example, this may be the case if you appear in an image shared by a user on our products or services or if someone mentions information about you in posts or captions shared on our products and services. »
What are the “AI experiments” that the data will be used for?
Meta already offers services in some countries such as the Meta AI chatbot, based on its Llama language processing model. The group also offers an image generation tool or gadgets allowing, for example, “to create personalized stickers for chats or stories”.
But user data could be used to develop other programs. And this is also one of the problems raised by the NOYB association. “Meta does not say for what purposes it will use this datadeplores Max Schrems, the co-founder of the association, in a press release. It could be for a simple chatbot, for very aggressive ad targeting, or even a killer drone. » The activist believes that this contravenes the European data protection regulation (GDPR).
Read the interview: Article reserved for our subscribers Max Schrems: “The GDPR is influential because it is and will continue to be the global standard”
Add to your selections
How to prevent Meta from exploiting its data for AI?
By default, from June 26, any user of Meta’s products will see their data used to feed their AI systems, unless they make their opposition known, a possible option for European users. Facebook provides a link for this, which can be found on the “privacy policy” page. But at the time of writing, it was not working.
On Instagram, you have to go to this page, which however suggests that any request will not necessarily be taken into account (“if your request is accepted, it will be applied in the future”). You must then fill out a form giving some basic information (country, email address) and, more surprisingly, justify your request: “Please tell us how this treatment will affect you. » To confirm your approach, you must then go to your email inbox to retrieve a code sent by Meta. And wait for the response, which can arrive within a minute by email.
” A joke “, denounces the NOYB association. According to Max Schrems, Meta “took care to design tons of little distractions to ensure that only a very small number of users bothered to object”.
Can the NOYB procedure against Meta succeed?
Max Schrems has in any case obtained several major victories in recent years against the digital giants. In 2015, the Austrian activist invalidated the “Safe Harbor” agreement, which regulated the way in which American companies could transfer the data of European citizens to the United States. In 2023, he also contributed to Meta being fined a record 1.2 billion euros under the GDPR.
The world
Special offer for students and teachers
Access all our content unlimitedly from €9.99/month instead of €11.99.
Subscribe
He is not the only one to be outraged: Internet users have expressed their anger on social networks and artists, who have been publishing their work on Instagram for years, have left the platform. On June 6, the specialist site TechCrunch reported that the recent Cara platform, founded by anti-AI artist Jingna Zhang to offer an alternative to Instagram with an emphasis on protecting user rights, had become a week from 40,000 to 650,000 users, following Meta’s announcements.
Read the decryption | Article reserved for our subscribers How artists try to defend themselves against artificial intelligence
Add to your selections
){kind=link}