Privacy and Ia

Meta's data collection to train AI starts today, but you can still oppose it

From today, comments, photos and public content on Meta's social media will be used for Meta Ai training. What changes for those who objected by Monday 26 May

by Chiara Ricciolini

27 May 2025Aggiornato alle 14.15 del 27 maggio

2' min read

2' min read

As of today, Tuesday 27 May, Meta started using data from Facebook, Instagram and Threads to train its artificial intelligence model. This includes data from photos, posts, comments and public interactions.

It is still possible to fill out the appropriate form to prevent Zuckerberg's company from using your data for training, but it is no longer possible to prevent the retroactive use of previously published content (that was possible until Monday 26 May): even if you choose to object, the refusal only applies to data shared from today onwards.

What data is collected

The collection only covers public posts and comments, as well as interactions with Meta AI. Meta does not use private messages exchanged with friends or family members to train its artificial intelligence models.

Public data from accounts of children under the age of 18 resident in the European Union are excluded from collection. Content posted on private profiles is also not used, unless they are comments left on public posts.

All communications protected by end-to-end encryption, such as chats on WhatsApp and Messenger, are not accessible to the company for data collection.

What is the opt-out model

Meta has chosen an opt-out consent model: users' personal data are automatically used for artificial intelligence training, unless they explicitly object. An approach that has been strongly criticised by several European data protection authorities, according to which the General Data Protection Regulation (GDPR) instead requires explicit and prior consent for such invasive processing. In other words, an opt-in model should be adopted, with a clear request for authorisation before the user starts using the platform, avoiding the need for the user to take steps to seek and revoke consent.

The Austrian organisation Noyb, founded by activist and lawyer Max Schrems, has sent a formal notice to Meta: in the letter, the company is urged to change its approach and obtain explicit consent for the use of data for artificial intelligence purposes. If it fails to do so, Noyb says it is ready to file an injunction or start a class action in court.(https://noyb.eu/en/noyb-sends-meta-cease-and-desist-letter-over-ai-training-european-class-action-potential-next-step)

How to oppose

To prevent Meta AI from using their personal data, users must fill out a form available in the Facebook Help Centre or in Instagram settings. Each request is evaluated individually, and Meta can decide to reject it, as required by the European regulation.

On Facebook, it is necessary to open one's profile, access the Privacy Centre and click on 'opt out'.

On Instagram, you go to Settings, then Privacy, then More Information and Support, then Information, Privacy Policy, and select 'right to object'.

Entering a reason is optional.