Meta has removed AI usage metrics from employee performance reviews, shifting focus back to overall impact while continuing to test a new agentic AI tool called Hatch.
Key facts
- •Meta's new performance guidance replaces AI usage criteria with language stating outcomes can be supported by AI or other means.
- •The company previously saw the rise of an internal leaderboard that tracked employee token consumption and labeled users as "Token Legends."
- •Hatch is an agentic AI tool currently being tested by employees on corporate devices ahead of a planned public release.
- •Some employees express privacy concerns regarding Hatch, citing a previous Meta project that tracked keystrokes to train AI systems.
- •Meta CEO Mark Zuckerberg has stated that no further mass layoffs are expected for the remainder of the year.
Meta has formally ended an incentive program that evaluated employees based on their use of AI tools. Internal guidance issued this week removes references to "AI Native" designations and token counts in performance reviews, shifting the focus back to employee contributions. While the company is easing pressure on AI usage, employees continue to test a new agentic AI tool known as Hatch.
Shift in Performance Evaluation
Previously, Meta employees were evaluated on their "AI-driven impact," with usage metrics leading to labels such as "AI Native" or "AI First." This approach faced criticism, including a lawsuit filed in July by employees who alleged that workers on leave were unfairly penalized for failing to meet usage targets. Meta spokesperson Tracy Clayton stated that the company evaluates employees based on their contributions and that labels like "AI Native" were never used for performance evaluations.
Introduction of Hatch AI
Meta is currently encouraging employees to test Hatch, an agentic AI tool capable of browsing the web and operating applications autonomously. While usage is not mandatory, some employees have begun using the tool for personal tasks like booking appointments. However, some staff remain hesitant to grant the tool access to personal accounts due to privacy concerns and past experiences with company projects that tracked keystrokes to train AI systems.
Impact on Workforce and Layoffs
Employees report that while they are no longer pressured to "tokenmax"—or artificially inflate AI usage metrics—Hatch consumes significant computing resources. Concerns persist among staff regarding potential future layoffs if AI tools successfully boost productivity. Although reports suggested Meta considered further job cuts following a May layoff of 8,000 employees, CEO Mark Zuckerberg has stated that no further mass layoffs are expected this year.
Timeline
- AprilAn employee-created leaderboard tracking AI usage was taken down after details were leaked.
- MayMeta laid off 8,000 employees.
- JulyA lawsuit was filed by employees alleging Meta violated antidiscrimination laws regarding AI usage metrics.
Advertisement
This article was independently rewritten by ManyPress editorial AI from reporting originally published by Wired.
