Meta pauses employee surveillance tool after privacy concerns over AI training

Meta has temporarily halted an internal employee activity tracking tool that was being used to collect data for training its artificial intelligence models. The move follows backlash from staff and privacy advocates who argued that the monitoring, which recorded details of workers’ digital behavior, was implemented without meaningful consent.

The decision was first reported by The Guardian. According to the report, the tool gathered data on how employees interacted with internal systems—such as which documents they accessed, how they communicated, and how they spent their time. Meta intended to use that data to improve its AI systems, but critics say the practice blurred the line between legitimate productivity monitoring and invasive surveillance.

What happened

The tracker, which had been running for some time, was paused after employees raised concerns. Reports indicate that workers were not clearly informed that their day-to-day interactions with company tools would be fed into AI training pipelines. Once the extent of the data collection became known, internal pushback grew, and Meta chose to stop the practice while reviewing its approach.

Privacy advocates have pointed out that in many jurisdictions, employee consent for data collection is legally ambiguous. Even when company policies allow monitoring, using that data to train AI models—especially models that may later be deployed publicly—raises questions about fairness, transparency, and the right to refuse.

Why it matters

This episode is not an isolated one. As companies race to build more capable AI systems, employee data has become a tempting source of training material. Other firms, including Google DeepMind and various tech giants, have also faced scrutiny over how they use workplace information for AI development. A recent report backed by the Trades Union Congress (TUC) in the UK called for workers to have a greater say in how AI is rolled out, particularly when it involves monitoring or data collection.

The broader concern is that employees often have little practical choice. Opting out can be impossible if the monitoring is built into the tools they must use. And once data is collected, there is rarely a guarantee that it will not be repurposed—for AI training or other uses—without additional consent.

Meta’s pause may signal that companies are beginning to recognize the reputational and legal risks, but it does not resolve the underlying tension between workplace efficiency and individual privacy.

What readers can do

If you are a worker concerned about your own data being used in AI training, there are a few practical steps you can consider:

  • Review your employer’s privacy and data policies. Many companies have acceptable use policies that describe what data is collected from company devices and networks. Look for language about “machine learning” or “model training.”
  • Separate personal and work accounts. Avoid logging into personal services like social media, email, or messaging on work devices. This reduces the chance that personal communications become part of a corporate data set.
  • Use encrypted personal communication channels for sensitive conversations. If your employer’s monitoring extends to messaging apps, consider using Signal or other end-to-end encrypted services on your own device for private discussions.
  • Ask questions. If you are unsure what data is being collected, ask your IT department or HR. In some regions, you may have a legal right to information about how your data is processed.
  • Support collective action. Unions and worker advocacy groups are increasingly pushing for clearer rules around workplace AI. If you have a union, raise the issue. If not, consider organizing with colleagues to demand transparency.

It is important to note that none of these steps offer perfect protection. In many workplaces, the terms of employment already grant broad surveillance powers. But being informed can help you make better decisions about what you share and where.

Sources

  • The Guardian, “Meta pauses employee tracker for AI training amid privacy concerns” (24 June 2026)
  • The Guardian, “Workers need greater say over AI rollout, says TUC-backed report” (29 May 2026)