Meta Platforms (META) is apparently weighing a sweeping workforce shake-up, according to a scathing new report.
According to a Seeking Alpha report citing Reuters, Meta executives are exploring layoffs that could affect 20% or more of the company’s workforce, though the plans have not been finalized and no timeline has been set.
However, in responding to the report, Meta spokesperson Andy Stone said those claims amount to “speculative reporting about theoretical approaches.”
Be that as it may, the context surrounding the cuts shows why the conversation is getting heated now.
Meta and other tech giants have been ramping up their AI investments.
The Facebook parent’s official guidance in its latest earnings report showed 2026 capital expenditures in the $115 billion and $135 billion range, per Investing.com.
Meta said the incredible spending bump is driven primarily by “increased investment to support our Meta Superintelligence Labs efforts and core business.”
On top of that, it forecast total expenses for 2026 of $162 billion to $169 billion, primarily linked to higher infrastructure costs and AI talent hiring.
If these layoffs materialize, they could affect more than 15,000 workers, considering Meta had 79,000 employees at the end of last year.
The AI boom is starting to reshape the tech workforce
If these layoffs go through, the rationale would be very different from that of typical cost-cutting layoffs.
Instead of sluggish demand or slowing advertising markets, the conversation’s about AI-driven efficiency.
That dynamic has been playing out with Meta over the past several months.
Jan. 12, 2026 — Reality Labs: Meta planned to chop 10% of the unit, equating to 1,500 jobs, linked to metaverse and VR projects.
Oct. 22, 2025 — AI teams: Around 600 roles were cut across FAIR, product AI, and AI infrastructure as Meta effectively reshaped the division.
Apr. 24, 2025 — Reality Labs: Meta laid off an undisclosed number of Oculus Studios staff as part of broader efficiency efforts. Source: Reuters
In fact, we could go back to its “Year of Efficiency” restructuring, when it cut more than 21,000 jobs across two rounds in 2022 and 2023, while also scrapping 5,000 open roles.
It’s important to note that Meta isn’t alone in laying off workers in the AI space.
Amazon confirmed 16,000 corporate job cuts in January, and then another round impacting its robotics division in March.
Similarly, other tech companies have followed a similar playbook.
Autodesk announced it would cut 7% of staff, Pinterest trimmed less than 15% of its workforce, and Atlassian announced 1,600 layoffs as it looks to go deeper into AI-driven products and enterprise software.
Taken collectively, it’s clear that in the tech sector, companies are looking to reshape their workforces in funding infrastructure, talent, and the computing power needed for the AI era.
Meta executives are weighing major workforce adjustments while pushing aggressively into costly artificial intelligence initiatives.Morris/Bloomberg via Getty Images ·Morris/Bloomberg via Getty Images
The latest jobs report shows subtle cracks in tech hiring
Headline payrolls: The February 2026 U.S. jobs report, released March 6, showed total nonfarm payrolls dropping by 92,000, a much softer print following a 126,000 increase in January.
Headline unemployment: The unemployment rate held at 4.4%, with 7.6 million unemployed, while labor-force participation was 62% and the employment-population ratio was 59.3%, underscoring a labor market that continues to cool off.
Data point supporting the AI layoff narrative: The clearest BLS datapoint that supports the AI layoff discussion is that information-sector employment dropped by 11,000 in February and has been declining by an average of 5,000 jobs per month over the past year. Source: U.S. Bureau of Labor Statistics
A separate data point from another independent layoffs report from Challenger, Gray, & Christmas also shows that the tech labor market might still be under a ton of pressure.
Tech pain stayed elevated in February: Challenger said U.S. employers announced 48,307 job cuts in February, down 55% from January and down 72% from a year earlier, but the tech space still logged a worrying 11,039 cuts in February and 33,330 year to date, up 51% from the prior-year period.
AI is no longer a side note in layoff language: Employers linked AI to 4,680 job cuts in February, which equates to 10% of all cuts confirmed that month. Year to date, AI was responsible for 12,304 cuts, or roughly 8% of announced layoff plans.
The hiring side makes the signal stronger: Though February hiring plans jumped from January, employers planned just 18,061 planned hires so far in 2026, down a massive 56% from the same period last year. Source: Challenger, Gray, & Christmas
AI is taking tasks faster than it is taking whole jobs
A lot has been made about AI displacing workers across the board, but as I’ve covered of late, the reality is far less clear-cut than many suggest.
A huge chunk of the evidence we’re seeing points more to task compression and a re-imagining of jobs, instead of a large-scale job replacement.
For instance, Indeed’s 2025 AI-at-work report showed that just 26% of U.S. jobs will be highly transformed by AI, while the lion’s share, 54%, are only moderately impacted, according to Hiring Lab.
Moreover, the report also showed that nearly 46% of skills in a usual U.S. job posting are poised for “hybrid transformation,” where human involvement still matters.
That’s supported by AI bellwether Anthropic’s real-world usage data. Per its Economic Index report, 36% of jobs show AI utilization in at least a quarter of their tasks, but just about 4% of occupations show AI use across three-quarters of tasks. For the most part, AI usage leans more toward augmentation (57%) instead of automation (43%).
Naturally, the biggest names in the tech world are split.
Anthropic CEO Dario Amodei predicted last year that in 2025, AI could obliterate “half of all entry-level white-collar jobs” within five years, Fortune reported.
On the flip side, Microsoft CEO Satya Nadella framed AI as more of a productivity layer, while Nvidia CEO Jensen Huang took a counterview that AI “won’t destroy jobs,” saying it would expand use-cases in radiology and help with shortages in fields like nursing.
Another big reason for the lack of broad displacement is that many AI models still make a ton of basic mistakes for multiple unsupervised business workflows.
In OpenAI’s published evaluations, GPT-4.5 had a 37.1% hallucination rate on SimpleQA, while the o3 and o4-mini system cards reported hallucination rates of 51% and 79%, respectively.
Oscar-winning actor and director Ben Affleck pushed back on the AI hype during a recent appearance on The Joe Rogan Experience.
