Meta stops tracking employees to train AI due to privacy concerns

What happened?

Meta, the parent company of Facebook, Instagram, and WhatsApp, has halted a program that collected computer usage data from its employees to train its artificial intelligence models. According to the BBC, the program began just two months ago and tracked activities such as application usage, web browsing, and idle times. The measure was taken after an internal report warned of potential privacy violations against workers. This incident adds to a long list of data management controversies for Meta, which has been fined over 1.3 billion euros by the European Union for GDPR violations.

Why is this important?

This case highlights the delicate balance between the need for data to train AI systems and employees' privacy rights. Meta, which has heavily invested in AI to compete with companies like OpenAI and Google, now faces scrutiny over its internal practices. The decision to halt the program could set a precedent for other tech companies considering similar data collection methods. According to a Gartner study, 60% of large companies already monitor employee activity, but only 30% clearly disclose the use of that data for AI training. This case could accelerate regulation in this area.

Context and background

This is not the first time Meta has faced controversies over data use. The company has been fined multiple times for privacy violations, including the Cambridge Analytica scandal in 2018, which affected 87 million users. However, using employee data to train AI represents a new frontier. Unlike user data, employees have a more limited expectation of privacy at work, but laws still protect their personal information, such as GDPR in Europe and state laws in the U.S. like California's CCPA. In 2022, a German court ruled against a company that monitored employee productivity through software, establishing that explicit consent is necessary. This precedent could apply to Meta if employees decide to sue.

Consequences for Meta and the industry

The suspension of the program could delay Meta's AI efforts, as internal employee usage data is valuable for improving productivity models and virtual assistants. Meta has allocated over $30 billion in AI investments in 2024, according to its annual report, and any delay could affect its competitiveness against OpenAI, which has launched GPT-4o, and Google, which integrates Gemini into its products. Additionally, the company could face lawsuits from employees who felt surveilled without their consent. A similar case occurred in 2023 when Amazon was sued for monitoring its warehouse workers, resulting in a $5 million settlement. For the industry, this case underscores the need for clear policies and transparency when collecting worker data. Companies like Microsoft have already established internal guidelines requiring explicit consent to use employee data for AI.

What should readers know?

Workers at tech companies should be aware of their employers' policies on data collection. Although companies have the right to monitor the use of their equipment, collecting data for AI training may require explicit consent under privacy laws. It is recommended to review employment contracts and company privacy policies. In Spain, the Spanish Data Protection Agency (AEPD) has issued guidelines stating that monitoring must be proportionate and transparent. Additionally, employees can request access to the data collected about them and object to its use for non-essential purposes. This case should also alert regulators: the European Union is already working on the AI Act, which could include provisions on using employee data to train models.

"Employee privacy should not be sacrificed on the altar of AI innovation. This case is a wake-up call for all companies." — analyst at TheVortiq