Want to stop Meta from using your info to train its AI models? Here’s how to do it

If you fear your personal information is being used by Meta to train its AI models, you can now opt out by following these steps.

Meta has updated its website to include a new form that lets you control your data. (Image: Zohaib Ahmed/Indian Express)

Listen to this article

00:00

1x 1.5x 1.8x

Generative AI is a powerful technology that can create new content from existing data, such as text, images, and audio. Meta is one of the leading players in this field, developing large language models like Llama 2 that can produce realistic and diverse outputs.

However, Meta’s AI also raises some privacy concerns, as it relies on massive amounts of data from publicly available and licensed sources. This data can include personal information, such as names, contact details, and opinions, that belong to individuals who may not be aware of or consent to their data being used for AI training.

To address this issue, Meta has recently updated its Facebook help center website to include a new form that allows users to access, modify, or delete any third-party information that was used for generative AI model training. The form, titled “Generative AI Data Subject Rights,” refers to third-party information as data that is publicly available on the internet or from licensed sources. It’s available on the Facebook help center resource section and can be used to submit requests related to personal data.

Also read | From Google Duet to Snapchat Dreams, these are the latest generative AI app updates

How to delete your Facebook data used for training generative AI models

Regardless, if you want to delete some of the data that Facebook has collected from you and used to train its generative AI models, here are the steps you need to follow:

– Visit the “Generative AI Data Subject Rights” form on Meta’s privacy policy page about generative AI from https://www.facebook.com/privacy/genai
– Click on the link that says “Learn more and submit requests here.”
– Pick one of the three options that best matches your issue or objection. The first option lets you access, download, or correct any of your personal information that comes from third-party sources and is used for generative AI training. The second option lets you delete any of that information from those sources. The third option is for anything else that you want to raise.
– After choosing an option, you will have to pass a security check test.

That’s it, your request will be submitted.

Meta also said in a related blog post that it uses this kind of information to train effective models that can unlock new advancements in generative AI. The company added that it is committed to being transparent about the legal bases that it uses for processing this information and that it respects the rights of data subjects.

Unfortunately, the form does not cover any user data that is generated or collected on Meta-owned platforms, such as Facebook comments or Instagram photos. A Meta spokesperson assured, though, that the company’s newest Llama 2 open-source large language model “wasn’t trained on Meta user data, and we have not launched any Generative AI consumer features on our systems yet.”