Meta has halted a controversial internal program that monitored employee workstations to train its artificial intelligence models following a massive security breach. The initiative involved tracking highly granular data, including mouse movements, keystrokes, clicks, and real-time screen content. This intrusive data collection was intended to provide a rich dataset for AI development but ultimately exposed sensitive information to unauthorized internal access.
The program's suspension came after an internal engineer issued a security warning, revealing that the harvested data was accessible to any Meta employee. This leak highlights the inherent risks of workplace surveillance and the potential for large-scale data exposure when internal safeguards fail. While the project was designed to improve technical efficiency, the fallout has sparked significant concerns regarding privacy and the security of internal corporate infrastructure.
