Meta’s Secret Sauce: How Big Tech’s AI Ambitions Are Cooking Up Job Losses

Mark Zuckerberg’s leaked comment about studying employees to improve AI models has sparked widespread concern over job losses at big tech companies. During an April 30 all-hands meeting, the Meta CEO candidly remarked that his company is studying employees “to figure out how to make this all more effective.” This statement suggests that Meta is prioritizing efficiency over employee welfare.

The essence of Zuckerberg’s comment lies in his assertion that internal employees serve as training data for Meta’s superintelligence models. By harnessing their collective intelligence and work output, the company aims to accelerate its AI development without investing heavily in research and development. This approach has significant implications for the labor market, particularly at companies where employee churn is already a pressing concern.

Recent layoffs at Meta are a stark reminder of this reality. Employees were told to work from home due to “humanitarian reasons” only to receive termination emails the next day, with access cards deactivated. For shareholders, this development sends a clear signal about where Zuckerberg believes operating leverage comes from: fewer employees and a more efficient use of resources.

According to Jason Calacanis on his podcast, Meta’s approach may not be unique. Calacanis questioned whether Zuckerberg’s candor is the first honest signal that AI is producing real job loss at big tech companies, beyond the usual platitudes about efficiency and innovation. Co-host Lon Harris noted that Google, Amazon, and OpenAI are likely doing the same thing, but Zuckerberg’s public acknowledgment sets him apart.

The trend of using internal workflow data to improve foundation models has far-reaching implications. If this becomes a widespread practice, every hyperscaler’s labor base will become training data for its AI models. This raises concerns about job security and the future of work at big tech companies. Will employees continue to be seen as essential assets or mere resources to be exploited?

Meta’s recent financials offer some insight into this question. The company posted Q1 2026 EPS of $10.44, beating consensus estimates on revenue of $56.31 billion, up 33% year over year. Operating margin sits at 41%, and net income jumped 61% year over year to $26.77 billion. However, what does this success story mean for employees who are no longer part of the equation?

The labor market is already reeling from the effects of automation and AI adoption. With big tech companies like Meta prioritizing efficiency over employee welfare, it’s clear that job losses will only increase in the coming months. The industry’s emphasis on AI development has created a culture where employees are seen as mere resources to be exploited, rather than valued assets.

As Calacanis aptly put it, this is “a really bad look” for big tech companies. It’s time for them to rethink their approach and prioritize employee welfare alongside AI development. The future of work depends on it.

The era of big tech’s labor base being seen as training data for AI models has only just begun. As we move forward, it’s essential to consider the implications of this trend and advocate for policies that prioritize employee welfare alongside technological innovation.