Meta Platforms is adjusting its internal policy on employee data collection after widespread backlash over its plan to record workers’ computer activity for artificial intelligence training purposes.
In an internal memo, the company said it will introduce new controls allowing employees to temporarily pause data collection for up to 30 minutes and apply for exemptions from the system. The memo was written by Stéphane Kasriel, a vice president in Meta’s Superintelligence Labs unit, which focuses on building advanced AI models.
The initiative initially involved tracking employee mouse movements, keystrokes, clicks and other on-device activity on US-based work computers. The data was intended to help train AI systems designed to automate workplace tasks and support the company’s broader push into AI agents capable of performing complex functions.
However, the rollout triggered strong internal resistance. Employees raised concerns about privacy, the extent of surveillance on company devices, and the broader implications of turning workplace behaviour into training data for machine learning systems. Some staff reportedly criticised the approach, describing Meta as an “Employee Data Extraction Factory.”
Kasriel acknowledged these concerns in the memo, stating that while the company remains confident in its privacy safeguards, it had listened to employee feedback regarding data sensitivity, device usage, and the need for greater control over when tracking occurs. The company said the system had undergone multiple layers of internal risk review before deployment.
In addition to giving employees more control, Meta said it has introduced technical improvements aimed at reducing the software’s impact on device performance. Staff had complained that the tool was consuming significant system resources, including battery life, and in some cases increasing home internet usage.
A Meta spokesperson declined to comment further on the changes.
The tracking initiative was launched as part of a broader restructuring effort within Meta, which has been significantly increasing investment in artificial intelligence development. The company’s AI division is focused on building systems that can carry out tasks with minimal human input, a strategic priority that has reshaped internal operations.
The controversy has also drawn attention from outside the company. Analysts note that the move could complicate Meta’s regulatory position in regions such as the European Union, where data privacy rules are particularly strict and tech firms face ongoing scrutiny over how employee and user data is collected and used.
While Meta has not fully withdrawn the program, the introduction of new safeguards signals a partial retreat in response to internal pressure, highlighting growing tensions between rapid AI development and workplace privacy expectations.
