Scale AI, the well-known data-labeling company that attracted global attention after receiving a monumental $14 billion investment from Meta, has recently undertaken significant staffing reductions that have drawn concern throughout the technology sector. One of the most affected units is its Red Team, a specialized group primarily responsible for rigorously stress-testing artificial intelligence systems and identifying potentially harmful outcomes before the models are deployed at scale. According to accounts shared by multiple current and former employees who spoke anonymously due to the sensitivity of the issue, twelve contractors from this critical team received unexpected termination notices late on a Friday evening.
The communication, which was reviewed by Business Insider reporters, was striking in tone and content. The message informed the individuals that their contracts were being ended on the grounds that their job performance allegedly no longer met the standards ‘required of the role.’ However, former team members expressed skepticism about this explanation, characterizing the move instead as a disguised round of layoffs. Several even took to LinkedIn to reveal their dismissal publicly, framing it as part of broader challenges facing Scale AI following Meta’s acquisition of a substantial ownership stake.
In response to the growing scrutiny, Natalia Montalvo, who serves as a spokesperson for Scale AI, offered a statement attempting to mitigate speculate concerns. She clarified unequivocally that these terminations were not symptomatic of a wider restructuring initiative but rather involved only a very limited segment of the workforce. Montalvo emphasized that the twelve affected individuals were contractors categorized within the company’s broader contingent labor pool. She further stressed that they represented only a minor fraction of the Red Team overall. According to her, Scale AI continues to allocate resources toward red teaming operations and remains firmly committed to delivering such safety and reliability services to prominent clients in the artificial intelligence industry.
Despite this official narrative, several workers suggested that the environment within Scale AI shifted dramatically after Meta’s multibillion-dollar investment. For instance, two former Red Team members noted that the steady influx of assignments and safety evaluations that had once sustained their daily operations decreased significantly in the months following Meta’s financial backing. They contended that the reduction in work assignments forced skilled specialists—who had been recruited specifically for their expertise in identifying potential risks in AI systems—to instead spend time on menial or non-specialized duties, thereby undermining the very purpose of the Red Team.
The staffing structure at Scale AI further complicates the situation. A large proportion of the company’s personnel, including the dozen recently dismissed, are contracted through staffing intermediaries such as HireArt. The termination emails themselves came not directly from Scale but from HireArt, underscoring the precarious position of contractors in the highly volatile AI sector. These communications mentioned that the employees would receive a month of severance pay but clarified that accumulated paid time off would not receive financial compensation. For logistical matters, the email instructed departing contractors to use a prepaid shipping label in order to return company-issued hardware or access badges.
Adding another layer of concern, one former employee estimated that this round of cuts could represent approximately half of the Red Team’s active headcount, though Scale AI declined to confirm the exact figures. The partial dismantling of such a crucial safety-focused group arrives only months after the company enacted a substantial downsizing in July, which eliminated 14% of its full-time staff in addition to cutting roughly 500 contractors. That earlier decision was, according to internal emails, explained by executives as a consequence of overhiring and shifts within the market. However, former insiders pointed out that those cuts also followed the loss of major clients, most notably OpenAI, which withdrew its business after Meta purchased a near-majority stake in Scale and integrated the company under its strategic purview.
OpenAI had been one of Scale AI’s most significant customers, relying heavily on the Red Team’s capabilities. The departure of Willow Primack, former leader of Scale’s Red Team, further complicated matters—she left the company to join OpenAI shortly after Meta finalized its investment. This transition not only exacerbated internal instability but also suggested deeper uncertainty regarding the continuity of client relationships.
In summary, while Scale AI frames the latest dismissals as performance-based decisions affecting a very small contingent of contracted workers, the broader context paints a picture of an organization struggling to reconcile its workforce strategy with shifting external conditions. The aftermath of Meta’s $14 billion stake, the departure of major partners like OpenAI, a reallocation of highly skilled employees to jobs outside their specialization, and ongoing rounds of workforce reductions collectively hint at the broader challenges facing an artificial intelligence company operating under immense pressure from market dynamics, corporate realignment, and changing client expectations.
Sourse: https://www.businessinsider.com/scale-ai-makes-cuts-to-key-team-after-meta-investment-2025-9
