Today, many social networks use the content that users publish to train artificial intelligence (AI) models. This practice has generated concern among those who do not want their information to be used for this purpose.
Although some platforms allow you to deactivate this function, in others it is not so simple or the option does not exist at all. Below, we explain how you can limit or prevent your content from being used on each platform.
LinkedIn recently added an option for users to opt out of the use of their posts and other data to train AI models.
By default, this option is activated, which means that your texts, photos, videos and shared links are being used for this purpose.
To disable this option, follow these steps:
- Sign in to your LinkedIn account.
- Access the “Privacy and data” settings within Settings.
- Disable the “Use my data to train AI models that create content” option by moving the button to the “Off” position.
This process is simple and will allow you to prevent LinkedIn from using your content in its AI training.
Goal
Meta, which includes Facebook, Instagram, Threads, and WhatsApp, does not offer an easy way to disable the use of your public posts to train AI. But if you are in Spain (or any EU country) Meta does not use your data (or so they say).
But if you’re in a country without strict privacy regulations, like the United States, there are some alternatives to limit this practice.
Make your posts private
Meta does not use private posts to train AI. On Facebook and Instagram, you can change the privacy of your posts and groups to “Friends Only” or “Private” respectively.
- On Facebook, go to Privacy Settings and adjust who can see your posts.
- On Instagram, go to Settings > Privacyand make your account private.
Avoid interacting with Meta AI chatbots
Meta has included chatbots in its applications that learn from interactions with users. If you prefer that your conversations not be used to train AI, avoid using these tools or delete the conversations.
Unfortunately, if you post content publicly, there is no direct way to prevent Meta from using it for its AI models other than not posting or minimizing public interactions.
TikTok
In the case of the famous Chinese platform TikTok, the platform relies on public interaction to train its AI algorithms, using generative AI to process text, images and videos.
Being a mostly public platform, it does not offer a clear option to opt out of this training.
Although TikTok doesn’t allow you to completely disable AI use of your content, you can adjust your privacy settings to limit how your information is shared.
Go to Settings and privacy > Privacywithin the app and adjust preferences for sharing your content and interactions. This won’t completely prevent it from being used in model training, but it will help limit it.
In 2024, Reddit signed a $60 million deal with Google for its content to be used in the company’s AI models. There is also no option for users to opt out of this process.
Since Reddit is a public forum, everything you post can be crawled and used in AI models.
If you want to reduce the risk, avoid posting to public subreddits or change your profile visibility to private, although this does not guarantee that your content will not be used.
Tumblr
Tumblr continues to resist and, for a few years, has been owned by Automattic (the same company that owns WordPress).
If you’re still using Tumblr, you should know that your posts may be shared with third parties, including AI projects like OpenAI and MidJourney. However, Tumblr offers the option to prevent your content from being shared:
- Enter the settings of your blog.
- Go to the “Visibility” section and activate the “Do not share data and content with third parties” option for each of your blogs.
If you have multiple blogs on Tumblr, you will need to repeat this process for each of them.
X (Twitter)
X (formerly known as Twitter) uses public posts to train its AI chatbot, Grok. However, you can disable this option:
- Login to X.
- Go to Settings > Privacy and securityand look for the “Data Sharing and Personalization” section.
- Look for the option with Grok and press it.
- Deactivate the option that says: “Allow your posts, as well as your interactions…”.
If you have previously chatted with Grok, you can also delete the history of conversations with the chatbot, from the same screen.
Although these steps will prevent your posts from being used to train the chatbot, they will still be publicly visible and could be used by third parties.
YouTube
YouTube, owned by Google, uses both the videos and their generated subtitles to train various AI models, including Google Gemini. Currently, there is no option to opt out of this process.
However, you can adjust other privacy settings to limit how your personal information is used. To do this:
- Go to your Google account.
- Go to Data and privacy and adjust the settings according to your preferences.
Discord
In Discord, by default, user information is not used to train AI, unlike other platforms. However, if you want to ensure that your data is not used for this purpose, you can easily do so:
- Go to User Settings > Privacy and security.
- Disable the options “Use my data to improve Discord” and “Use my data to personalize my Discord experience”.
This will limit the use of your data to personalize the experience on the platform, although according to Discord’s terms and conditions, your messages are not used to train AIs.
Twitch
Amazon-owned Twitch also contributes to training AI models, although it does not offer a clear option to disable this feature.
While it is possible to adjust some privacy settings to limit the use of your data in other contexts, there is no direct way to prevent Twitch from using your content in AI training.
Why should I care about limiting how my content is used?
The reasons why users decide to disable the use of their content to train AI are many.
Many users do not want their personal data, photos, videos or interactions to be used for any purpose without their explicit consent. Privacy is a crucial and sensitive topic for many people.
Other users want to maintain control over how and where their content is used. They don’t like the idea of their posts, images or data being used by third parties without their approval.
There are many people who have doubts about the ethics of using personal data and user-generated content to train AI models. They may worry about the possibility of their data being used in ways they consider inappropriate or that they do not support.
Finally, there is a real fear that content could be manipulated or used maliciously. Deepfakes and other types of AI-generated content can be used to misinform, create deepfakes, or mislead others.
The best way to protect your content is to regularly review the privacy settings of each platform and decide what level of public interaction you are willing to accept.
This post is also available in: