Meta (META.US) CTO radical reform: using AI to reshape the employee system aims to create a "management-free" large team.

According to reports, Meta's Chief Technology Officer Andrew Bosworth is leading a new mission - using artificial intelligence (AI) to reshape the company's workforce system. The report states that Bosworth is known for being outspoken and driving Meta's technology direction, especially in the AI and infrastructure fields. It is reported that when employees request to opt-out of certain projects, he does not show any compromise or apologize. Regarding concerns from employees about privacy issues related to the monitoring of company devices, he suggested employees not use personal emails on company devices. According to previous reports, Meta is installing new tracking software on employees' computers in the United States to capture mouse movements, clicks, and keyboard inputs in order to train its AI models. An internal memo indicates that this is part of a larger plan at Meta to create AI agents that can autonomously perform work tasks. One memo shows that the tool, called the "Model Capability Plan," will run on work-related applications and websites, and will periodically capture content on employees' screens. This is aimed at improving Meta's AI models in simulating human-computer interactions, such as dropdown menu selections and keyboard shortcuts. Bosworth told employees in a memo that the company will enhance internal data collection as part of its "AI for Work" program, now renamed the "Agent Transformation Accelerator." Currently, Bosworth is leading Meta's efforts in the next-generation computing platform and consumer hardware field, including projects such as Quest, Horizon, and the Meta Glass product line. In an internal memo, Bosworth told employees, "As we expand our efforts to build AI agents that can replace employees in completing tasks, our roles will shift to guiding, auditing, and helping them improve." He also stated, "We have seen tasks that used to take hours now only take minutes. Soon, on some tasks, we won't even need to be personally involved." He hopes to build large teams with few or no managers. Additionally, Bosworth plans to eliminate traditional planning documents and written proposal processes, preferring teams to directly build and test runnable prototypes. Last week, Meta officially began a 10% global workforce reduction (involving around 8,000 people) as part of its restructuring efforts to improve efficiency, reduce costs, and increase AI investments. In addition to direct layoffs, Meta will undergo disruptive organizational changes, implementing a flatter management structure by significantly reducing managerial positions. Meta's Chief Human Resources Officer, Jennell Gale, pointed out in a memo that many department leaders are introducing "AI-native design principles" in designing the new architecture, and the company will operate in the future with leaner, more autonomous small team models to achieve faster response times. In this restructuring known as "comprehensive AI transformation," in addition to facing unemployment, Meta also plans to transfer up to 7,000 employees to new projects related to AI workflows, with some transitioning already underway. Combining layoffs and job transfers, this restructuring will directly impact approximately 20% of Meta's workforce, while the company has also closed an additional 6,000 open positions. Meta's series of actions are primarily aimed at saving on labor costs amidst a surge in AI capital spending. On April 29, Meta released its first-quarter financial report, with revenue up 33% year-on-year to $56.31 billion, with advertising revenue up 33% year-on-year to $55.02 billion; operating profit at $22.87 billion, up 30% year-on-year. The company expects revenue in the second quarter to be between $58 billion and $61 billion. At the same time, Meta has raised its full-year 2026 capital expenditure forecast to $125 billion to $145 billion, with both the upper and lower limits increased by $10 billion from before. Meta stated in its financial report that the increase in capital expenditure forecast is mainly due to expected price increases in components this year, as well as additional data center costs to support future capacity. Despite substantial revenue growth driven by AI-driven precision targeting, the company continues to streamline its workforce to "offset other investments." Gale labeled the layoffs in the memo as an action to "improve company efficiency," stating plainly that this is to "offset the other investments we are making." It is worth noting that this is not the first time Meta has undergone large-scale workforce reductions. Since laying off around 11,000 employees in November 2022, Meta has conducted multiple rounds of layoffs, with the action involving around 8,000 people being the largest adjustment since the "year of efficiency" in 2023.