After returning to the AI gaming table, is Zuckerberg's first move layoffs?

Original | Odaily Planet Daily (@OdailyChina)

Author｜Azuma (@azuma_eth)

Reuters reported exclusively on April 18 that three informed sources revealed that Mark Zuckerberg’s Meta plans to launch the first round of large-scale layoffs this year on May 20, with further layoffs to follow.

One insider said that Meta will cut about 10% of its global staff in the first round of layoffs (total employees approximately 79,000), roughly 8,000 people. Another source indicated that Meta also plans to further reduce staff in the second half of this year, but the specific timing and scale have not yet been finalized, as ongoing observation of AI capabilities development may lead Meta executives to adjust the plan.

In a previous Reuters report last month, insiders also revealed that Meta was considering layoffs of 20% or more.

As of the time of writing, Meta declined to comment on the timing and scale of the layoffs.

10 days ago, Meta just caught up with the AI wave

Only 10 days ago, Meta’s AI development team, “Superintelligence Labs” (Meta Superintelligence Labs, MSL), led by the highly recruited Chinese-American genius Alexandr Wang, just released its first self-developed AI model Muse Spark.

Alexandr Wang disclosed that over the past nine months, MSL rebuilt the entire AI tech stack from scratch. Muse Spark is a native multimodal reasoning model supporting tool invocation, visual chain of thought, and multi-agent orchestration. This is the most powerful model Meta has released to date. During training, MSL observed that the model showed predictable scalable improvements in pretraining, reinforcement learning, and reasoning during testing.

Muse Spark also supports a “Contemplating Mode,” which orchestrates multiple parallel reasoning agents, specifically designed to handle complex scientific problems and reasoning tasks. In testing, MSL found its performance competitive with extreme reasoning models like Gemini Deep Think and GPT Pro.

As Meta’s first substantial product after heavily investing in AI and shifting to a closed-source model, Muse Spark is widely regarded by the market as the beginning of Meta’s pursuit of top-tier AI companies like Anthropic, OpenAI, and Google. Although Meta admits that this model still lags behind the flagship models of the top three companies in some capabilities, for Zuckerberg, who has long fallen behind in AI competitions due to the failure of the Llama route, Muse Spark and subsequent models in the series are enough as chips for him to re-enter the AI arena.

Market signals for Muse Spark are also positive. On the day of release, Meta’s stock closed at $612.42, up 6.5%, and over the following 10 days, it continued to rise (also influenced by the overall market rally), with yesterday’s closing price reaching $688.55.

The sword of AI has first fallen on employees

From late 2022 to early 2023, Meta launched the controversial “Year of Efficiency” plan, which involved the largest layoffs in the company’s history, cutting about 21,000 jobs. This time, it is very likely to be Meta’s largest round of layoffs since the “Year of Efficiency.”

Compared to the “Year of Efficiency,” Meta faced huge stock price declines and adjustment pressures after excessive growth during the pandemic. Now, Meta’s financial situation is clearly more stable, but the management’s vision is — a leaner organizational structure with fewer management layers, aided by AI to achieve higher efficiency.

Business Insider reported last month that, based on leaked internal documents, Meta is actively encouraging employees to use AI tools more proactively, with a target that — by mid-2026, 65% of engineers will have over 75% of their code written with AI participation.

According to the self-media Official Layoff (@LayoffAI), which focuses on layoffs at large tech companies (source unverified), “Starting this year, Meta has incorporated ‘AI-driven influence’ into all employee performance evaluations, making it a core metric. Without using AI, promotion is impossible. Meta has become the first large tech company to officially link AI usage to promotions.”

AI white-collar iteration, no longer an isolated case

Using “AI iteration to boost productivity” as a reason for layoffs is no longer an isolated phenomenon.

Last October, Amazon laid off as many as 30k employees across logistics, payments, gaming, and cloud computing departments. CEO Andy Jassy had previously hinted at this round of layoffs: “As the company increasingly uses AI to perform tasks previously done by humans, Amazon’s workforce may shrink.”

At the end of February this year, Jack Dorsey’s fintech company Block announced 4,000 layoffs, reducing its staff from over 10k to less than 6,000, to promote a leaner, flatter, AI-centric organizational structure. Block’s CFO and COO Amrita Ahuja revealed that after the layoffs, many corporate executives proactively contacted Block seeking to replicate this “script.”

• Odaily note: see “Jack Dorsey’s company, 4,000 white-collar workers being eliminated by AI.”*

Earlier this week, Meta’s core product Instagram’s direct competitor Snap also cut about 1,000 jobs. CEO Evan Spiegel stated: “AI will enable our team to reduce repetitive work, improve efficiency, and better support our community, partners, and advertisers.”

Now, the same wind has blown to Menlo Park, California, where Zuckerberg has raised his sword.

Oh, and one more thing worth mentioning. Although Jack Dorsey once publicly claimed that “the rapid development of AI is iterating the traditional productivity growth paradigm,” shortly after Block’s layoffs, many of the dismissed employees received rehire invitations (see “The first batch of big tech layoffs by AI have already returned to work”)…

AI white-collar iteration may eventually become a reality, but rushing to cut 40% of staff like Block did can easily “go too far and backfire.”