English
What Is Meta's Keystroke Monitoring Program?
Meta, the parent company of Facebook, Instagram, and WhatsApp, has launched a controversial initiative called the Model Capability Initiative (MCI) that tracks every keystroke, mouse movement, click, and periodic screenshot on U.S. employees' work computers. The data is being used to train Meta's artificial intelligence models to replicate human-computer interactions, according to internal documents obtained by multiple news outlets including The New York Times and CNBC.
The monitoring software, deployed by Meta Superintelligence Labs (MSL), captures employee activity across hundreds of websites and applications, including Google, LinkedIn, Wikipedia, GitHub, Slack, and Atlassian. Meta's own platforms like Threads and Manus are also included. The company states the program aims to build AI agents capable of performing complex office tasks by learning from real-world examples of how people use computers.
Employee Backlash and Internal Unrest
Employees have reacted with fury to the tracking program. Internal correspondence seen by The New York Times reveals hundreds of staffers responded with angry emojis, while many voiced concerns openly. One employee wrote: 'Your indifference to the concerns of your own employees is alarming.' Another called the initiative 'incredibly demoralizing.'
The backlash is amplified by Meta's announcement that it will lay off 10 percent of its workforce — approximately 7,800 to 8,000 employees — effective May 20, 2026. The Meta layoffs 2026 impact has left many workers fearful of speaking out, as they await notification of whether their positions are affected. Chief People Officer Janelle Gale confirmed the cuts are part of Meta's shift toward AI-driven roles and operational efficiency.
No Opt-Out Option
Meta's CTO Andrew Bosworth confirmed in an internal meeting that there is no opt-out option for employees using company-issued devices. This announcement triggered a wave of negative reactions on internal forums. The top-voted comment on one thread simply asked how to opt out, drawing hundreds of angry reactions.
European employees are reportedly exempt from the monitoring due to the General Data Protection Regulation (GDPR), which imposes strict limits on workplace surveillance. This has further fueled resentment among U.S. staff, who feel their privacy is being sacrificed for Meta's AI ambitions.
Meta's Justification: AI Training, Not Surveillance
Meta executives have sought to calm the storm. CEO Mark Zuckerberg addressed the concerns in an internal meeting, emphasizing that the data collection is not intended to spy on employees or evaluate performance. 'The data is meant to train our AI on how smart people use their computers to perform their tasks,' Zuckerberg said, according to attendees.
A Meta spokesperson issued a statement confirming that safeguards are in place to protect sensitive data, though the company declined to provide specifics. The spokesperson reiterated that the collected data is used exclusively for AI training and no other purpose. Meta also noted that employee activity on company machines has been monitored in some capacity for years, and the MCI program simply expands on existing practices.
What Data Is Being Collected?
- Every keystroke typed on company-issued work computers
- Mouse movements and click locations
- Periodic screenshots of work-related applications
- Activity across hundreds of pre-approved websites and apps including Gmail, Google Chat, VS Code, Slack, GitHub, and LinkedIn
Meta insists that incidental personal data captured on screen — such as passwords or personal messages — will not be learned by the AI models. However, privacy experts have expressed skepticism about the effectiveness of such safeguards.
Privacy and Ethical Concerns
The initiative has drawn sharp criticism from privacy advocates and legal experts. In the United States, there is no federal privacy law that specifically limits workplace surveillance, making Meta's program technically legal. However, in Europe, the same program would likely violate GDPR, which requires explicit consent and limits the extent of employee monitoring.
Dr. Sarah Chen, a privacy researcher at the Electronic Frontier Foundation, told reporters: 'This level of surveillance — every keystroke, every mouse movement — creates a chilling effect on workers. Even if Meta promises the data is only for AI training, the potential for mission creep is enormous.'
Some employees fear the data will be used to train AI agents that could eventually replace them, especially given the concurrent layoffs. The AI replacing workers debate has intensified as Meta invests heavily in automation. The company has committed up to $135 billion in capital expenditure for 2026 on AI infrastructure, including a 49 percent stake in data-labeling firm Scale AI for over $14 billion.
Broader Industry Context
Meta is not alone in seeking workplace data for AI training. Several tech companies have been scouring internal communications — including Slack archives and Jira tickets — to feed their AI models. However, Meta's approach of real-time keystroke monitoring is unprecedented in its scope and intrusiveness.
The MCI program is part of CEO Mark Zuckerberg's push to catch up in generative AI, where Meta has lagged behind competitors like OpenAI and Google. The company recently launched its Muse Spark model, and the Meta Muse Spark AI model is expected to benefit from the employee data collected through MCI.
Impact on Meta's Workforce and Culture
The combination of mass layoffs and pervasive monitoring has created a toxic atmosphere at Meta, according to multiple employee reports. The New York Times described the mood as 'miserable,' with staff feeling both surveilled and disposable. One employee wrote in an internal post: 'The indifference to the concerns of your own employees is deeply troubling.'
With notifications about layoffs scheduled for May 20, many employees are waiting in limbo while their every digital move is tracked. The situation has led to comparisons with dystopian workplace surveillance, with some staff considering resignation.
Frequently Asked Questions
What is Meta's Model Capability Initiative (MCI)?
MCI is a Meta program that installs tracking software on U.S. employees' work computers to capture keystrokes, mouse movements, clicks, and periodic screenshots. The data is used to train AI models to perform office tasks autonomously.
Can Meta employees opt out of the monitoring?
No. Meta's CTO Andrew Bosworth confirmed there is no opt-out option for employees using company-issued devices. European employees are exempt due to GDPR protections.
Is Meta using the data for performance reviews?
Meta states emphatically that the collected data is used exclusively for AI training and not for performance evaluations or any other purpose.
How many employees is Meta laying off in 2026?
Meta is laying off approximately 10 percent of its workforce — about 7,800 to 8,000 employees — with notifications beginning May 20, 2026.
What safeguards does Meta have for sensitive data?
Meta says safeguards are in place to protect sensitive content, but the company has not provided specific details. Privacy experts remain skeptical about the effectiveness of these protections.
Follow Discussion