Intensified "K-shaped split" within Meta: elites earn millions in annual salaries, ordinary employees reduced to "AI fodder"

Meta is conducting a Silicon Valley experiment in class stratification. As artificial intelligence reshapes the organizational structure of big tech companies, a new fault line is deepening within the enterprise—the AI elite enjoy sky-high salaries and unlimited resources, while other employees face the shadow of layoffs, morale collapse, and some even hope to be included in the layoff lists.

This phenomenon is known as the “K-shaped company”—much like the logic of wealth disparity in the K-shaped economy, companies are forming two sharply different tracks: AI elites at the top see their salaries and status rise, while ordinary employees at the bottom are increasingly viewed as replaceable parts.

Meta employees have openly criticized management, circulated petitions, and even attempted to form unions in the UK.

Management has publicly acknowledged issues with these practices and begun to roll out a series of appeasement measures—but analysts point out that structural imbalance cannot be bridged simply by improving employee benefits.

CEO Mark Zuckerberg recently admitted that the company’s AI agent development progress has not met expectations; Meta’s newly launched Muse Spark AI system also failed to match OpenAI and Anthropic’s cutting-edge models in key benchmark tests.

The AI Elite’s Feast: Nine-Figure Salaries and Special Benefits

Within Meta, the treatment enjoyed by top AI talent is worlds apart from ordinary employees.

Top AI executives and researchers receive compensation plans as high as nine figures (hundreds of millions of dollars), with some even offered up to $500,000 in additional equity incentives simply for staying on—while clouds of mass layoffs loom over other departments.

One Meta employee bluntly stated, "The only ones truly doing well are those with the highest pay and the closest ties to AI development." Another was even more direct: "The only people not dissatisfied are the executives."

This internal stratification is not unique to Meta, but reflects Silicon Valley's long-standing dual-track employment structure—the gap between full-time workers and contractors has existed for years. The novel factor brought by the AI wave is that even full-time employees risk marginalization if they aren’t in the core AI circle.

The Dilemma of Ordinary Employees: Surveillance, Layoffs, and Morale Collapse

For those at the lower rung of the K-shaped structure, circumstances are drastically different. Some are forced into what insiders describe as a “suffocating” team; headcount is constantly reduced, and even as company profits grow, median pay is falling.

More controversially, the company tracked employees’ keyboard and mouse activity to train AI agents—a project halted only after encountering security breaches which employees had previously warned about.

Chief Technology Officer Andrew Bosworth bluntly summarized management’s overall stance: "You can leave, or choose to stay with dissent but still execute."

Low morale has become a management crisis Meta can no longer ignore. In the latest round of possible layoffs, some employees even hope to be laid off—an irony for Meta, once one of the most coveted employers in corporate America.

Harvard Business School professor Amy Edmondson said in an interview: "They’ve confused Silicon Valley’s ‘fail fast’ credo with reckless execution. This isn’t the inevitable chaos that comes with AI innovation, but a fundamental management failure."

AI Progress Fails to Meet Expectations, Management Trust Overdrawn

Internal turmoil has begun to negatively affect Meta’s core strategic goals. Zuckerberg recently admitted that Meta’s AI agent development speed is lagging behind executive expectations; the newly released Muse Spark AI system falls behind OpenAI and Anthropic’s main models in terms of performance metrics, and the company faces difficulties in developing open-source AI models.

Management has also recognized the problem. Executives privately described the launch of the company’s new AI organizational structure as "absolutely terrible," and admitted that the work environment has become "brutal."

When Zuckerberg recently announced a hackathon to boost morale, employees reportedly reacted with indifference—not only are they exhausted, but one remarked: "I’m not sure this company still supports hackathon culture. I don’t feel secure enough to dedicate time to innovation unrelated to performance metrics."

Chief Technology Officer Bosworth gave an unusually candid diagnosis in an internal memo: "We’ve damaged your trust—you believed your professional abilities and contributions would be valued, believed you could grow and advance, believed this is truly a place where you can make an impact."

The Cost of Fission and Uncertain Recovery

Currently, Meta has launched a series of measures to placate employees, including promises of more attentive management support, greater stability, no more mass layoffs this year, increased team budgets, and improvements to office infrastructure. A widely criticized unassigned seating policy is also being scaled back.

But Laszlo Bock believes this may indicate a deeper turning point—the continuous erosion of employee voice in the post-pandemic era may now be in a “green shoot” stage of recovery. Meta’s story shows that viewing employees as interchangeable parts rather than partners ultimately harms the business.

The core paradox of this K-shaped stratification is that while AI should flatten organizational structures, it has created even more rigid power hierarchies in reality. How to drive technological leaps while maintaining basic internal trust is now a management issue large tech companies cannot avoid.

Risk Warnings and DisclaimerThe market involves risks, and investment should be cautious. This article does not constitute personal investment advice and does not consider individual users’ particular investment objectives, financial situations, or needs. Users should consider whether any opinions, views, or conclusions herein are suitable for their specific circumstances. Investing accordingly is at your own risk.