X is the latest social media site letting 3rd parties use your data to train AI models

X to Share User Data with AI Firms Starting Nov. 15

Elon Musk’s X, the platform formerly known as Twitter, has been using user data to train its artificial intelligence models. Soon, other companies will gain access to that data for similar purposes.

Starting November 15, X will begin sharing user information—including posts, likes, bookmarks, and reposts—with third-party platforms, which may use it to train AI models.

The social media company announced the changes through an update to its privacy policy on Wednesday. When the policy takes effect, users will be automatically opted in unless they manually opt out.

"Depending on your settings, or if you decide to share your data, we may share or disclose your information with third parties," the policy states. It adds, "If you do not opt out, in some instances the recipients of the information may use it for their own independent purposes... including, for example, to train their artificial intelligence models, whether generative or otherwise."

As user data grows in value, platforms are seizing the opportunity to profit from it by licensing their information to AI companies. "This is the latest arms race. Everyone is working towards AI supremacy," said Toronto-based cybersecurity and tech analyst Ritesh Kotak. "The more data sets you have, and the more people from whom that data is collected, the more accurate your model will be."

A Trend Among Social Platforms

The policy change follows X’s earlier shift, which allowed the platform to train its in-house Grok chatbot on user data. However, that decision triggered an investigation by the European Union’s privacy regulator, resulting in X halting data collection from the region for training purposes.

X isn’t the only platform to adopt these practices. LinkedIn uses user data to train its AI models, and Meta relies on public Instagram and Facebook posts to build AI tools. Platforms like Reddit, Stack Overflow, and Tumblr have also struck lucrative deals with AI companies, signaling a growing trend in the industry.

Ajay Shrestha, a computer science professor at Vancouver Island University, noted that social media companies are exploring new revenue streams amid declining ad revenues. "The traditional processes—whether through advertising or subscriptions—are not working well," Shrestha said.

Examples of such deals include:

• Reddit signed a $60 million annual content-licensing agreement with Google.
• Stack Overflow began charging AI companies for data scraping last year.
• Tumblr and WordPress reached agreements with Midjourney and OpenAI to sell user data for AI training.

Some publishers and stock image companies have also benefited from similar arrangements. Shutterstock, for example, generated over $100 million in licensing revenue last year. However, many others have pursued legal action against AI firms for scraping their content without permission.

The Value of Social Media Data

Kotak emphasized that social media posts are especially valuable to AI companies because they capture emotions and reflect how people communicate. "Social media posts may not always contain high-quality information, but they are rich in sentiment analysis," he explained.

Can Users Opt Out?

As of Friday, X had not yet updated its settings to provide an opt-out option ahead of the policy change on November 15, according to CBC News.

Kotak cautioned that many users may not want their personal information used to train AI models. "These platforms making it a default setting that your data is used for training leaves you without a real choice—unless you go in and stop it," he said.

Typically, users can opt out by navigating to "Settings," selecting "Privacy and Safety," and toggling off the "data sharing with business partners" option. However, Kotak noted that opting out may not undo what’s already been done. "Once an AI model is trained on your data, there’s no way to reverse that or remove your information from the model," he said.

His parting thought underscores the reality of the situation: "If you’re not paying for the product, you are the product. And in this case, the data is the product."