Petition against Meta's employee training data collection for ML models
4 hours ago
- #employee rights
- #data privacy
- #AI ethics
- Employees demand that Meta not collect employee 'computer-use' data (mouse movements, keystrokes, screen content, etc.) for training AI models due to privacy, consent, and trust concerns.
- The 'MCI' program reportedly lacks completed privacy reviews and offers vague mitigations, with selective opt-out for executives, raising transparency and fairness issues.
- Past incidents, such as GDPR fines for plain-text password storage and an AI agent causing a data leak in 2026, underscore security and regulatory risks.
- Employee data is protected under laws like CCPA and CPRA, granting rights to know, delete, correct, and limit use of personal information.
- The letter argues that intrusive, non-consensual data collection contradicts Meta's stated commitment to 'Building AI Responsibly' and empowering individuals.
- Signed by over 1,000 employees, including engineers, managers, and researchers, calling for a clear commitment against such data collection.