Starting May 27, 2025, Meta, the tech giant owner of some of the world’s most used social platforms such as Facebook, Instagram, and WhatsApp, will implement a new policy that will allow the use of European users’ personal data to train its artificial intelligence (AI) system known as Meta AI. This decision by Meta, which applies exclusively to adult users within the European Union, has generated considerable controversy, not only because of the magnitude of the data that will be collected but also because Meta will rely on the concept of “legitimate interest” to access the information without requiring explicit consent.
That is to say, if Meta users do not actively object, the company will assume they accept the use of their personal data to improve its AI, which has raised concerns about privacy and the control users have over the information they share on these platforms. The massive use of personal data by Meta raises a series of challenges regarding transparency and users’ right to privacy.
Meta’s collection will include a wide variety of user-generated content, from photos, videos, and comments, to social interactions such as “likes” and reactions to posts. However, private messages, both on Facebook Messenger and WhatsApp, will be excluded from Meta’s collection.
Faced with this new measure, Meta users will need to make informed decisions about whether they want to allow their data to be used or if they prefer to exercise their right to object. In this article by ITD Consulting, we will address in detail what this Meta policy implies, what type of data will specifically be used, how users can object, and what are the possible implications for privacy in a world increasingly dominated by artificial intelligence.
What is Meta AI?
Meta AI is the artificial intelligence system that Meta has been developing and perfecting over the last few years. This Meta AI aims to create more personalized and interactive experiences for users through its social platforms, such as Facebook, Instagram, and WhatsApp.
Meta AI has multiple applications: from virtual assistants that answer questions, to image creation, content generation, and advanced analysis of user behavior. Meta’s AI requires large volumes of data to train its models. This data helps improve the accuracy of machine learning algorithms and allows Meta’s system to be more efficient in tasks such as behavior prediction, content recommendation, or automated response creation.
In this context, Meta has decided that the data generated by users on its platforms will be essential to train its AI. This includes a vast amount of personal information generated daily on social networks, such as photos, posts, interactions, and comments.
Meta’s Announcement: Why does Meta want to use your data?
In April 2025, Meta announced that, starting May 27, 2025, it would begin using the personal data of European users to train its artificial intelligence. This Meta measure is based on what the company describes as “legitimate interest,” a legal concept under which companies can use users’ personal data without asking for explicit consent, as long as the user does not actively object.
In other words, if you do not take action to object, Meta will assume you accept that your data be used to train its artificial intelligence. This “opt-out” practice that Meta has adopted has been criticized by privacy organizations and many users, who consider it unfair that their data be used without their explicit consent.
What data will Meta use to train Meta AI?
The key to this controversy lies in the type of data Meta plans to use to train its artificial intelligence. Although Meta has assured that certain data, such as private messages or information from minors under 18 years old, will not be used, adult users who share content on platforms like Facebook and Instagram will see their data included in Meta AI’s training. Below, we detail what type of data Meta will use:
1. Photos and videos
Meta will use the photos and videos you upload to Facebook and Instagram to train its AI. This includes all types of visual content you upload, whether as posts, stories, profile photos, or any other images shared on your Meta accounts. These images will serve to train Meta AI in tasks such as object, person, place, and emotion recognition.
2. Comments and posts
All comments you leave on posts, as well as your reactions to other Meta users’ posts, will be used to train the AI. This type of data is fundamental for Meta AI to learn about human behavior and social interactions. Comments help train natural language processing models, allowing Meta AI to better understand the meaning behind words, emotional nuances, and users’ intentions.
3. Stories and real-time reactions
Instagram stories and real-time interactive posts will also be part of the data used to train Meta AI. Real-time interactions, such as reactions to posts or live comments, help Meta AI understand immediate responses and how conversations and emotions develop over time.
4. Captions and descriptions
Every time you add a caption or description to an image, Meta may also use that information. Written descriptions provide additional context that helps Meta AI understand the visual content of images and posts. This is also key to training content generation systems that can learn to automatically create captions and descriptions.
5. Interactions with other users
The way you interact with other users, from giving “likes” to sharing posts or reacting to comments, will also be used by Meta. These interactions provide valuable information about users’ interests and behaviors, helping Meta AI predict what might interest a person based on their previous interactions.
What will Meta not use?
Although Meta plans to use almost all publicly generated data on its platforms, there are some exceptions. These are the areas excluded from Meta’s AI training process:
1. Private messages
Meta has assured that it will not use private messages you send through Facebook Messenger or WhatsApp to train its AI. Direct messages between Meta users will be outside the data collection, providing a certain layer of privacy. However, this point could change in the future if Meta decides to include them in its models, which could raise even more privacy concerns.
2. Data from minors under 18 years old
Another key aspect of this policy is that Meta has clarified it will not use data from minors under 18 years old. This includes any type of content generated on platforms by users who have not reached the age of majority. Generally, European data protection law is designed to protect minors’ personal information, and Meta has respected this restriction.
Criticism of Meta’s Policy: An Unfair Exploitation of Personal Data?
The use of users’ personal data to train Meta’s artificial intelligence has generated a series of criticisms. The main concern is that, by activating this process by default, Meta is taking advantage of users’ lack of information or laziness to change privacy settings. Many Meta users may not be aware that their data is being used in this way or may not know how to exercise their right to object.
Furthermore, the massive collection of user data raises serious questions about privacy. To what extent are users aware of the amount of data they are sharing on their platforms? Although Meta promises that private messages will not be used, many users do not understand the implications of having their posts, photos, and comments used to train AI models without explicit consent.
On the other hand, the way Meta has structured the policy has also been seen as an attempt to “capture” users who do not take action to object. Instead of requiring affirmative consent, Meta assumes users agree if they do not say otherwise, which could be seen as a strategy to facilitate data collection without active user intervention.
How to Object to Meta Using Your Data?
If you do not want Meta to use your personal data to train its AI, you have the right to object. However, the process to do so varies slightly across Meta’s different platforms.
On Instagram:
- Access the Instagram app: Log into your account and go to your profile.
- Go to Settings and Activity: Tap the three horizontal lines icon in the top right corner and select “Settings and Activity.”
- More information and help: Scroll down until you find this option and tap on “Information.”
- Privacy Policy: Enter this document and look for the link that says “Object.”
- Objection form: Complete the form that appears. Provide your email address and, if you wish, you can add a brief explanation about why you do not want your data to be used.
- Confirmation: Once the form is submitted, you will receive an email confirming that your request has been processed and that Meta will respect your decision.
On Facebook:
- Access Facebook: Log into your Facebook account and go to the user menu.
- Privacy Center: Tap on “Privacy Center” and look for the section related to generative AI.
- Objection form: Similar to Instagram, you will need to complete a form where you can explain your objection.
- Confirmation: You will receive an email confirming that Meta will respect your decision.
WhatsApp and Other Methods:
At the moment, WhatsApp is not included in the data collection for training Meta’s AI, but this could change in the future. You can also use external platforms such as Citizen8, which allows you to automatically send an email to Meta to object to the use of your data.
Why is Meta Doing This?
Meta states that using users’ personal data is essential to improve the quality of its artificial intelligence. By using data generated by millions of users on its platforms, Meta can create more advanced and useful AI models. These models will not only improve Meta’s internal tools but could also be licensed to other companies to use Meta’s AI in their own products.
Additionally, massive data collection is a fundamental strategy for Meta in its fight to remain competitive against tech giants like Google, Amazon, and Microsoft, which are also investing large amounts of resources in AI development.
Impact on Privacy: A New Challenge for Users
Massive data collection to train artificial intelligence poses serious risks in terms of privacy. As social platforms become omnipresent elements in our lives, our personal data is being used in ways that many users do not understand or anticipate. This type of use of personal data has much broader implications, from controlling our information to how our online interactions are interpreted and used by automated systems.
The widespread trend toward using data to improve artificial intelligence and personalize user experiences online must be accompanied by a debate about privacy. Users need to be informed and aware of what is happening with their data. In this sense, Meta’s “opt-out” policies are not sufficient to protect users’ privacy.
Meta’s policy to use European users’ personal data to train its AI marks an important milestone in the relationship between social networks and artificial intelligence. Although it is presented as a way to improve user experience and service quality, it raises a series of challenges for privacy and consent.
The key for users is to be informed about how their data is being used and how they can exercise their rights. As artificial intelligence continues to advance, concerns about online privacy will only increase, so it is essential that tech companies are transparent and respect users’ rights in this new digital paradigm.
Over time, we will likely see more regulations, policies, and security measures attempting to balance the advancement of artificial intelligence with the protection of personal privacy. The fight for privacy will be one of the greatest battles in the next decade, and how companies like Meta manage our data will be key in the digital future.
If you want to stay informed about the latest in technology and how Meta advances with new policies in favor of its AI, write to us at [email protected]. We have a specialized technology team to help you stay at the forefront.
EN 
ES