Social Media Platforms Are Using Your Data to Train AI Models: Here’s How to Opt Out

As artificial intelligence (AI) becomes more integrated into various technologies, major social media platforms are tapping into user data to improve their AI models. While some platforms offer ways to opt out, users should understand that publicly shared content can still be accessed by third parties for AI training.

Several prominent platforms are using data to enhance their AI capabilities:

• OpenAI has acknowledged using copyrighted materials, including conversational data, to train models like ChatGPT.
• LinkedIn utilizes user resumes to refine its AI systems.
• Snapchat employs selfies for personalized ads.

This practice has raised concerns among users about transparency and consent. David Ogiste, founder of Nobody’s Cafe, highlighted the ambiguity around how platforms use data for AI. He called for clearer opt-out options and more openness to alleviate user fears about AI.

How Platforms Handle User Data for AI Training

Here’s how some major platforms approach data use for AI models and user privacy:

• LinkedIn: Allows users to opt-out of data sharing for AI training, though previously collected data may have already been used.
• X (formerly Twitter): Users must manually opt-out if they don’t want their posts included in training data for Grok, the platform’s AI.
• Snapchat: Uses selfies to generate targeted ads with AI but offers an opt-out option.
• Reddit: Shares public user data for AI training with third parties, but private data remains protected.
• Meta (Facebook and Instagram): Public posts are used for AI, but private messages are excluded.

While opting out is possible on some platforms, it’s essential to recognize that any publicly shared content can still be used by third parties for AI development. Understanding how your data is used and knowing how to opt out can help you retain some control over your personal information.