This is a large-scale move that has sparked backlash and concern over privacy protection.
Meta, the parent company of Facebook and Instagram, is preparing to use public data from adult European users of these social networks as training material for its advanced artificial intelligence models.
It is a large-scale initiative that has triggered reactions and concerns about privacy, despite the legal justification cited by the company under the “legitimate interest” clause of the General Data Protection Regulation (GDPR).
Users now essentially hold the responsibility for protecting their digital privacy. However, they have limited time to act, as failure to object means their posts may be used as training data. Private messages, underage accounts, and WhatsApp conversations are currently excluded.
The Austrian NGO NOYB has submitted complaints in nine countries, while courts such as the one in Cologne have shown an ambiguous stance, leaving room for further legal developments.
Meta is developing its own chatbot: Meta AI
Meta is working on a new artificial intelligence model named Meta AI — a chatbot capable of answering questions, generating text, and even creating images based on descriptions. It will, for example, assist with writing an email, planning an outing, or producing illustrations. It is positioned as a direct competitor to ChatGPT and Google Gemini, and it will be powered by data from users’ social media content.
The raw material
To develop Meta AI, the company will use information shared by users, such as their name, profile picture, and posts in public Facebook groups or on public Instagram profiles. Content from private groups or private profiles will not be included. According to Meta, this data is necessary to help its AI better understand European languages and cultures.
Obstacles from European regulators
The American company officially announced its intentions on 14 April. It had planned to launch this initiative earlier but was forced to delay last year due to concerns from European regulators, who argued that users had not been given enough time to object. Now the plan is back on track, and although the Dutch Data Protection Authority is still assessing its legality, no bans have been enforced.
How can users object?
Users must fill out a specific request form through their Facebook or Instagram settings. On Instagram, this is done via the Privacy Centre on the user’s profile, and on Facebook, the process is similar through Privacy Settings. The form is linked within a paragraph that explicitly mentions the option to object.
The process applies to content published by the user who completes the request. For photos or texts shared by third parties, a second, more complex form must be filled out. In such cases, Meta requests additional details about the published content and the reason for the objection.
Deadline for objection
The date of 27 May is not a strict cut-off: it will still be possible to submit an objection afterwards, but by then, part of the user’s data may already have been collected.
Objections will not have retroactive effect, and there is a risk of partial exposure. Meta acknowledges this in its privacy policy: “We may continue to process information about you for the development and improvement of Meta’s AI services even if you object or do not use our products.” This includes posts made by third parties.
Legal coverage and protection measures
Meta invokes “legitimate interest”, a legal basis provided under the GDPR. This stance was endorsed in December 2024 by the European Data Protection Board, despite concerns from several privacy organisations. The Austrian NGO NOYB has filed complaints in eleven European countries. However, the Cologne Court of Appeal in Germany recently judged that Meta’s plan does not breach the relevant legislation.
France’s Data Protection Authority (CNIL) reminds that invoking legitimate interest is only permissible if three conditions are met: the processing must have a lawful purpose, be necessary, and must not disproportionately infringe on the rights of those concerned. To a large extent, responsibility now falls on users to react swiftly — otherwise, their posts may be used to feed the algorithms of the American tech giant.
Remaining options
Those wishing to further assert their digital rights can do so, but it is a time-consuming process. In any case, users can also send an email to formally express their refusal to allow their data to be used, while also hoping that the EU’s plans to reform data protection rules will soon be set in motion.
Also read: Meta adds teen restrictions to Facebook and Messenger
Source: Premium Edition of TA NEA – Economy Today
