He presses
Without the consent of its users, Meta Group plans to use private data to train artificial intelligence. However, action can be taken against this.
FRANKFURT – A change in guidelines on Instagram and Facebook is currently causing a stir. From the end of June, users' private data from social networks will be fed to artificial intelligence (AI). Parent company Meta recently paused its project in Europe due to concerns from data and consumer advocates. However, according to the Consumer Advice Center of North Rhine-Westphalia (NRW), it is still worth filing an objection.
Private data from Instagram and Facebook is fed to AI: Meta receives a warning
According to its own information, Meta originally planned to use publicly available online data as well as worldwide content “shared across Meta products and services” as of June 26 to train its AI. According to the New South Wales Consumer Advice Centre, this is what explains the use of generative AI. According to the group, these are specifically “their posts, photos, and captions as well as the messages they send to the AI.”
What does AI do in Meta?
Specifically, artificial intelligence is called generators that use commands to generate, i.e. generate, texts, images and videos within a few seconds. Well-known and freely available applications include Chat-GPT, Dall-E, or Copilot. Now the Internet company Meta wants to follow suit and develop its own artificial intelligence that responds to commands. In order to do this, Meta needs as much data as possible from its users for “training purposes.”
(Source: NRW Consumer Center)
Meta is based on legitimate interest in the use of User Content and refers to Article 6 of the General Data Protection Regulation (GDPR). The only way to take action against the project is to file an objection. That's why the US group has been warned by the Consumer Advice Centre. This believes that users must consent first before sending their private Instagram and Facebook data to the AI.
Want to protect your data on Instagram and Facebook? This is how you can object
Moreover, the contradiction is very complex and does not always work, criticize consumer advocates. Meta provides a form called “Submit a Dispute” on Instagram and Facebook. However, this is only displayed when you are logged in. Users then have to explain why their data should not be used to train the AI. Whether the reasons given by those affected are sufficient will then be examined on a case-by-case basis.
This is how it works on Instagram:
- Log in to your account.
- Go to settings.”
- You can access the data protection guidelines via the “Information” menu item.
- Click on the “Right to object” link in the upper text field.
This is how it works on Facebook:
- Log in to your account.
- Go to Settings & Privacy.
- You can access data protection guidelines via the Settings menu item.
- Click on the “Right to object” link in the upper text field.
Data on Instagram and Facebook for training purposes: This conflicts with the Meta Plan
In order for Meta to agree to the objection, it is necessary to provide valid arguments in the text field “Please tell us how this processing affects you”. The North Rhine-Westphalia Consumer Center offers some suggestions, including:
- I hold the copyright to my published data and do not grant any right to use the AI applications.
- I fear that my published statements will be republished in the wrong context.
- I generally feel uncomfortable with the idea of artificial intelligence.
However, affected users reported to consumer advocates that Meta accepted their objections shortly after submitting the form — without them having to provide a lengthy explanation. However, interception should not always work, for example due to lack of access, for example through a hacked account. In some cases, there are also said to be technical issues that unnecessarily complicate the process. (All of them)
