## AI Chatbots Transmitting User Data to Meta, TikTok and Google — Research Documents Sharing Despite Privacy Settings
Major AI chatbot platforms are forwarding user conversations to third-party advertising trackers embedded in their interfaces, according to research released this week. The study examined ChatGPT, Claude, Grok, and Perplexity, finding that each service transmitted data to external companies including Meta, TikTok, and Google — including scenarios where users had declined cookie tracking.

Researchers tested the platforms across multiple privacy configurations and consent states. The findings showed consistent transmission of user prompts, session metadata, and device identifiers to ad networks operating within chatbot pages. The practice persisted regardless of whether users had selected opt-out settings, raising questions about the effectiveness of current consent mechanisms in AI products.

Privacy researchers point to embedded third-party code as the structural cause, noting that ad networks can capture information through tracking pixels and scripts loaded alongside chatbot interfaces. The findings add scrutiny to how AI companies monetize free or low-cost products and whether disclosed data practices align with what users experience. Industry observers say the research could accelerate regulatory attention on AI-specific privacy frameworks, particularly in jurisdictions where chatbot data collection remains only partially addressed by existing rules. The companies behind the chatbots had not issued public responses at time of publication.
---
- **Source**: Decrypt
- **Sector**: The Lab
- **Tags**: AI, Privacy, Data Leakage, Third-Party Trackers, Ad Networks
- **Credibility**: unverified
- **Published**: 2026-05-07 23:31:38
- **ID**: 80346
- **URL**: https://whisperx.ai/en/intel/80346