New research has found that Facebook, WhatsApp, and TikTok are the most privacy-invasive social media platforms, while Discord ranks as the safest for users regarding data privacy practices and AI training.
The Social Media Privacy Ranking 2025, conducted by Incogni, evaluated the data handling and privacy practices of 15 major social media platforms. The analysis reflects a significant shift compared to last year, when Reddit, Snapchat, and Pinterest were considered most privacy-invasive. According to the latest findings, Meta’s platforms as well as TikTok now lead the ranking for privacy risks.
Privacy and AI data risks
The study examined the privacy policies of Discord, Pinterest, Quora, Telegram, Snapchat, Twitch, LinkedIn, Reddit, X (formerly Twitter), YouTube, Facebook Messenger, Instagram, TikTok, WhatsApp, and Facebook. Researchers assessed their practices regarding AI, user consent, transparency, regulatory penalties, and the handling of sensitive data.
The report concludes that 12 of the 15 platforms surveyed may use or allow personal user data to be used to train AI models. Only Telegram, Twitch, and Discord clearly indicated that user data would not be used for AI training purposes. This suggests that a large majority-approximately 80%-of social media companies falling under scrutiny might be utilising user information to develop or improve artificial intelligence systems.
Meta’s suite of products, which includes Facebook, Instagram, and WhatsApp, as well as TikTok, were found to be the most privacy-invasive, receiving penalties across all assessment categories.
Discord top for privacy
Discord was rated as the least privacy-invasive platform. A significant factor in Discord’s favour was its explicit stance against using user data for generative AI training, distinguishing it from competitors. Pinterest and Quora followed Discord in offering comparatively better privacy protection, according to the study’s scoring model.
The ranking marks a notable change from last year’s edition, which had seen Reddit leading on privacy, trailed by Snapchat and Pinterest. The declining positions of these platforms in 2025’s analysis are attributed mainly to changes in their policies around artificial intelligence and user data processing.
Meta faces regulatory scrutiny
The new assessment also highlights regulatory actions against large social media companies for breaches of privacy law. Facebook received the most fines, including one fine in the United States, four fines under the European Union’s General Data Protection Regulation (GDPR), and five additional fines in other countries, marking it as the most penalised among the platforms reviewed.
The study further reports that some platforms collect sensitive demographic data. LinkedIn is noted for possibly collecting users’ race or ethnicity data. Meanwhile, Meta’s platforms (excluding WhatsApp) and LinkedIn may collect data about users’ sexual orientation and health, expanding the types of sensitive personal information being handled by these companies.
Differences among platforms
Platforms such as YouTube, Snapchat, Pinterest, X, and LinkedIn have suggested in their privacy documentation that user information may be used to generate or enhance AI models. In contrast, few companies provide clear assurances that personal data will not be used for such purposes.
“Social media users have the right to know where and how their personal information is being used, especially given the rise of data breaches and cybercrime in recent years,” said Darius Belejevas, Head of Incogni. “This year’s rankings reveal sharp contrasts in data privacy practices among platforms, demonstrating that we must continue to fight for greater transparency, stronger safeguards, and enhanced user control over sensitive information.”
Researchers believe that the combination of evolving AI practices and lack of transparency raises concerns about the adequacy of data protection on widely-used platforms.
The report concludes that greater efforts are required to safeguard privacy and maintain user confidence, as major platforms continue to process and potentially share increasingly sensitive personal data.
