Meta has paused its internal “Model Capability Initiative” (MCI), a program that tracked employee computer activity to train artificial intelligence models, following widespread internal backlash and concerns regarding data security. The company confirmed to the BBC that the initiative is currently suspended while the company investigates how some collected data was left accessible to internal staff.
Why did Meta pause the tracking program?
Meta halted the initiative on Monday after discovering that internal data—which included keystrokes and mouse movements—was potentially accessible to unauthorized employees within the company. A Meta spokesperson stated that while there is no current evidence of improper data access, the program remains on hold to ensure proper security protocols are met.
The program, which launched approximately two months ago, was designed to gather behavioral data to improve Meta’s AI development. However, it faced immediate resistance from staff who cited privacy concerns and a lack of transparency regarding how the sensitive information would be protected.
How did employees respond to the surveillance?
The suspension follows weeks of intense pressure from the company’s workforce. Nearly 2,000 Meta employees signed a petition demanding the cancellation of the MCI program, according to reporting by the BBC.
In an attempt to mitigate the friction, management initially offered employees the option to opt out of tracking for 30-minute intervals. Staff members characterized this gesture as “damage control,” noting that the lack of initial consent for the surveillance had already significantly damaged internal morale. This tension coincides with a broader period of instability at the firm, marked by significant layoffs and a massive pivot toward AI-focused projects, which the company expects to cost up to $145 billion in 2024, as noted in recent financial filings.
The growing tension over AI-driven restructuring
The fallout from the MCI program reflects a deeper frustration among staff regarding Meta’s aggressive shift toward AI. While many technical workers acknowledge the need to compete with industry rivals like OpenAI and Anthropic, the execution of these changes has caused friction.
Reports from outlets including Wired have highlighted instances of employees openly questioning or insulting management during internal meetings regarding the company’s direction. For some, the rapid reorganization of teams and the intense focus on AI feels like a reactive strategy rather than a sustainable path forward. As Meta continues to pour billions into its AI infrastructure, the balance between technological advancement and employee relations remains a critical challenge for leadership.
Timeline of the Model Capability Initiative
- Two months ago: Meta launches the MCI program to collect data for AI model training.
- September 2024: Internal petitions reach nearly 2,000 signatures from employees protesting the surveillance.
- September 23, 2024: Meta officially pauses the program following reports of potential data accessibility issues.
- Ongoing: The company continues to conduct an internal investigation into the security of the collected information.