Meta Pauses Employee Tracker for AI Training: What You Need to Know

What happened

According to a June 25 report from The Guardian, Meta has temporarily stopped using an internal employee monitoring tool that was collecting behavioral data for AI model training. The move came after privacy concerns were raised both inside the company and by outside observers. The tool, which tracked how workers interacted with internal systems, was intended to improve Meta’s AI training datasets. But critics argued that using employee data without explicit, informed consent crossed a line—especially when those workers had little choice about participating.

Meta has not said how long the pause will last or what changes might follow. The company did confirm that the tracker is suspended while it reviews its data collection practices.

Why it matters for users and workers

This incident is not just an internal HR story. It highlights two broader issues that affect anyone who uses Meta’s platforms—Facebook, Instagram, WhatsApp, or Messenger.

First, workplace privacy is fragile. Many companies already monitor employee activity—keystrokes, screen time, email patterns. Meta’s case shows that such data can be repurposed for AI training without clear consent. If your employer uses similar tools, your own behavioral data might be used in ways you didn’t expect. This isn’t hypothetical: a growing number of companies are exploring ways to feed internal data into large language models or other AI systems.

Second, Meta’s appetite for training data is enormous and often controversial. The employee tracker pause follows a separate story from early May 2026, when New Mexico proposed a $3.7bn fine against Meta for alleged privacy violations on its social platforms. That case centers on how Meta collects and uses user data for advertising and other purposes, including AI development. Together, these events suggest that Meta is under increasing pressure to rethink how it obtains consent—whether from employees or from users.

For everyday consumers, the key question is: if Meta is willing to track its own people for AI training, what does it do with your data? The company’s terms of service allow it to use posts, messages (in some cases), and interactions for training its AI models, unless you opt out. Most users don’t realize that option exists.

What you can do to protect your privacy

You cannot control what Meta does inside its own offices, but you can take steps to limit how your personal data is used for AI training on Facebook and Instagram. These settings are not front and center, but they are available.

  • On Facebook: Go to Settings & Privacy → Settings → Privacy → Your Facebook Information. Look for a section called AI Training or Data for AI (the label may change over time). You should see an option to “Share your Facebook data with Meta to train AI models” or similar. Turn it off.
  • On Instagram: The process is similar. Navigate to Settings → Privacy → Data Sharing with Meta → AI Training. Toggle the setting to off.
  • For WhatsApp: Meta has separate policies for WhatsApp. Currently, WhatsApp does not use your messages for AI training in most regions, but the company has indicated it may change this. Review WhatsApp’s privacy policy periodically and consider using alternative apps if you want stronger guarantees.
  • Check your account privacy settings monthly. Meta often changes the location of privacy controls. Bookmark the direct link to the “Data Use” section if possible.

It’s worth noting that turning off these settings does not erase data already used for training, but it stops future collection for that purpose. Many privacy advocates recommend doing this even if you don’t plan to leave Meta’s platforms—it’s a small but meaningful step.

Broader context: not just Meta

The Meta employee tracker story is part of a larger conversation about consent in AI training. In May 2026, Google DeepMind began talks with UK unions over staff concerns about how its AI tools are used by the US and Israeli governments. Labour MP Dawn Butler recently sued Elon Musk’s xAI over fake sexualized images generated by Grok, highlighting that AI data sourcing and content pollution are overlapping problems.

What these cases share is a common thread: data used to train AI is often collected without meaningful consent. Whether it’s employee monitoring, social media posts, or public web scraping, the gap between what companies collect and what users expect is widening. Regulators are starting to take notice—New Mexico’s proposed fine and the EU’s ongoing Digital Services Act investigations are two signs of that shift.

What to watch next

Meta’s pause is likely temporary, but it may lead to real policy changes. The company could introduce clearer consent procedures for employees, or it might abandon internal monitoring for AI training altogether. For users, the bigger risk is that Meta and other tech companies continue to expand the scope of data they pull into AI models without clear disclosure. Watch for:

  • Updates to Meta’s privacy policy in the coming months.
  • Regulatory actions in the US (New Mexico case, potential FTC involvement) and Europe.
  • Whether other companies like Google, Microsoft, or Apple follow similar paths with internal data.

For now, the takeaway is practical: review your privacy settings, understand what you’re agreeing to, and remember that if a company is willing to track its own employees, it’s probably treating your data similarly.

Sources:

  • The Guardian, “Meta pauses employee tracker for AI training amid privacy concerns,” June 25, 2026.
  • The Guardian, “New Mexico proposes $3.7bn fine for Meta and sweeping changes to its social platforms,” May 6, 2026.
  • The Guardian, “Google DeepMind in talks with UK unions amid staff concern over US and Israel’s AI use,” May 20, 2026.
  • Meta privacy settings pages (accessed June 2026).