Meta to turn employee behaviour into AI training data, fueling privacy concerns

Meta, the prominent social media platform, is taking a controversial step by introducing employee monitoring software on workplace computers. This initiative aims to track various activities such as mouse movements, clicks, and keystrokes, as reported by Reuters. The data collected will serve to enhance the company's artificial intelligence (AI) models, enabling the creation of AI agents capable of performing work tasks independently. The internal tool, known as the Model Capability Initiative (MCI), will operate on specific applications and websites that employees frequently use. It will also capture occasional screenshots to provide context about their activities. In an internal memo shared by an AI research scientist from the Meta SuperIntelligence Labs, employees were informed about this development. Andrew Bosworth, Meta’s Chief Technology Officer, communicated to staff that the company is intensifying its internal data collection efforts as part of its AI for Work (AI4W) initiative. While the internal training of AI could prove beneficial, it raises significant concerns regarding employee privacy. Although Meta has not detailed what specific types of data will be utilized, Bosworth reassured employees that the information gathered through the MCI will not influence performance evaluations or other workplace decisions. Additionally, the company claims to have measures in place to safeguard sensitive information, though it remains unclear what constitutes 'sensitive content.' Bosworth emphasized the collaborative role of employees in enhancing AI models, stating, "This is where all Meta employees can help our models get better simply by doing their daily work." He elaborated that to develop agents that assist with everyday computing tasks, the models require authentic examples of user interactions, including mouse movements and button clicks. As Meta pushes to integrate AI into routine operations, it joins a growing list of tech firms adopting AI tools for task automation. CEO Mark Zuckerberg is advocating for the development of consumer-oriented tools that address real-world issues. With competitors like Anthropic and OpenAI also focusing on enterprise AI solutions, the ripple effects of these advancements are felt across the tech industry, influencing both stock performance and employment dynamics. Amid this AI surge, Meta is reportedly planning significant layoffs beginning May 20, potentially impacting around 10% of its workforce, equating to approximately 8,000 employees.