Under the sharp increase in AI investment, Meta (META.US) is busy with a "blood transfusion": 7,000 people transfer to AI, while simultaneously laying off 8,000 people.

In the face of the dual pressure of the AI arms race and efficiency pressure from Wall Street, Meta Platforms Inc. (META.US) is staging a large-scale "blood exchange" of technology talent. A recent internal memo reveals that Meta is reallocating 7,000 employees to AI-related positions, including teams newly formed for AI engineering and Intelligent Agents Accelerator, to promote the company's transformation towards an AI-first organizational structure. At the same time, the company is expected to begin cutting approximately 8,000 employees globally starting this Wednesday (May 20th), accounting for about 10% of its total workforce. Meta CEO Mark Zuckerberg has established AI as the company's top strategic priority, driving internal adoption of AI intelligent agents to assist in tasks such as coding. Zuckerberg previously stated that now a single individual can accomplish projects that previously required large teams. The 7,000 employees being reassigned will join teams such as the Applied AI Engineering (AAI) and Intelligent Agents Accelerator. These projects, led by Meta's Chief Technology Officer Andrew Bosworth, aim to develop AI intelligent agents that can autonomously perform tasks currently done by humans. The Central Analytics team will provide support for the development of these intelligent agents. However, internal resistance is brewing during the transformation process. Employees have been expressing their protests through means such as posting flyers in company offices and posting on internal communication platforms. Over 1,000 employees have signed a petition opposing the installation of mouse-tracking software on employee computers for AI model training. In the UK, some employees have begun organizing union movements, accusing the management of pursuing a speculative AI strategy at the expense of employees facing destructive consequences. AI capital expenditure surges, making way for labor costs Prior to this restructuring, Meta had already stated in its Q4 2025 earnings report that AI-related investment for 2026 would increase from $115 billion to $135 billion. The company achieved record revenue of $56.31 billion in the first quarter, but still chose to streamline its workforce to "offset other investments." Chief People Officer Janelle Gale characterized the layoffs in the memo as an action to "increase company efficiency," stating plainly that this was to "offset other investments we are making." This is not Meta's first large-scale workforce reduction. Since laying off approximately 11,000 employees in November 2022, Meta has undergone multiple rounds of layoffs, with this action affecting about 8,000 people being the largest adjustment since the "efficiency year" of 2023. Meta is not alone in this trend. Microsoft Corporation recently introduced its first voluntary retirement program in its 51-year history, targeting U.S. employees with a combined age and tenure of 70 years or more. About 7% of U.S. employees are eligible, with a potential departure of around 9,000 people. Prior to this, Microsoft Corporation had already laid off over 15,000 employees in 2025. Meanwhile, its subsidiary LinkedIn announced another round of layoffs affecting approximately 5% or nearly 900 employees, impacting engineering, product, and marketing teams, despite the company's latest quarterly revenue showing a 12% year-over-year increase. Amazon.com, Inc.'s AI transformation is even more aggressive. Since early 2025, Amazon.com, Inc. has laid off over 30,000 corporate employees, with layoffs primarily focused on non-core business departments, while frontline positions in e-commerce delivery and AWS cloud services remain unaffected. Alongside the layoffs, Amazon.com, Inc. requires employees to follow an "AI-first" principle in almost all work processes, from coding to product design and supply chain analysis, and plans to invest around $200 billion in AI infrastructure by 2026. However, this push has sparked internal reactionsaccording to internal employees, immature AI tools have increased the workload of manual error correction. Over 100,000 employees have been laid off since 2026 According to statistics from Layoffs.fyi, over 100,000 employees worldwide have been laid off in the tech industry since 2026, approaching the level of approximately 124,000 for the entire year of 2025. In the first quarter of 2026, driven by productivity tool impacts and Wall Street pressure, the shift from manual labor to AI automation has become an irreversible structural trend. It is worth noting that this wave of layoffs is different from before. Quoting executive coach Anthony Tuggle, foreign media outlets point out that "this represents a fundamental structural shift, not a temporary market adjustment. We are witnessing a permanent transformation of work organization across industries." Tech companies are shifting budgets from labor to GPU chips and data centersthe four tech giants (Meta, Microsoft Corporation, Amazon.com, Inc., and Alphabet) are expected to spend nearly $700 billion on AI infrastructure, while salaries for traditional tech roles remain stable, roles such as AI engineers enjoy a premium of over 50%. Meanwhile, the recruitment landscape is being reshaped. Demand for entry-level and generic IT positions is slowing down, while there is strong demand for senior engineers with AI integration capabilities. As economist Daniel Zhao points out, many employees are feeling "between a rock and a hard place" because stability no longer depends on tenure but more on whether their skills align with the demands of the AI era.