Mass Layoffs in Silicon Valley Due to AI?
This article is a column, and its entire content and opinions represent the personal views of the author and do not reflect the editorial stance.
TUESDAY, 31 March 2026. Thousands of Oracle employees worldwide received an email—sent at six in the morning.
The content was brief: their contracts were ending. No warning. No meeting. The company’s shares, ironically, rose 4-6 percent on the same day.
That is the face of digital transformation in the AI era. Not a friendly face. The numbers are enticing yet alarming.
Oracle is cutting between 20,000 and 30,000 employees—about 18 percent of its 162,000 workforce.
The most affected units are Revenue and Health Sciences, SaaS and Virtual Operations Services, and NetSuite India Development Centre.
The lost positions include software engineers, account executives, programme managers, cloud engineers, solution consultants, and many more. Not just lower-level employees, but also the backbone of the company’s operations.
Why did these layoffs occur? There are at least three narratives circulating, each containing varying degrees of truth.
The first narrative: AI is directly replacing workers. Bloomberg has confirmed that some of the eliminated positions are in categories of jobs estimated to no longer be needed due to AI.
The second narrative: Layoffs are a way to fund AI infrastructure. Oracle is building massive data centres. It has signed a $300 billion US dollar contract with OpenAI and become a key partner in the $500 billion US dollar Stargate project alongside SoftBank and several technology partners such as Microsoft and NVIDIA.
Cutting operational salaries could generate cash flow savings of $8–10 billion US dollars per year, according to estimates from TD Cowen analysts.
The third narrative: Some analysts see Oracle shedding the legacy burden of the Covid-19 pandemic. Like many other technology companies, Oracle had inflated its employee numbers during 2020–2022. Some of the current layoffs may be a correction for that over-recruitment.
These three narratives are not mutually exclusive. They work together, reinforcing each other.
Oracle is not an anomaly. It is part of a large wave that has been ongoing since 2025 and is now reaching its peak.
Microsoft cut around 15,000 employees throughout 2025. Amazon laid off 16,000 corporate employees in January 2026, while continuing to pour more than $80 billion US dollars into AI investments.
Meta has been gradually cutting thousands of employees since 2025, and is now preparing reductions that could affect 20 percent of its 78,000 employees, to fund $135 billion US dollars in AI capital expenditure in 2026.
Block, the company behind Square and Cash App, cut 4,000 employees, nearly 40 percent of its entire workforce.
CEO Jack Dorsey was frank: not because of financial difficulties, but because of AI’s continuously developing capabilities to perform more tasks.
Globally, nearly 60,000 technology workers lost their jobs in the first three months of 2026, equivalent to about 700 people per day, in an industry that is simultaneously posting record revenues.
If this trend continues, total layoffs could reach 265,000 people throughout the year, surpassing the 245,953 people from the entire year of 2025.
What makes this phenomenon different from previous layoff eras is the striking paradox: these companies are not dying.
Oracle recorded a net profit of $6.13 billion US dollars in the last quarter, up $95 US dollars. Amazon recorded revenue of $716.9 billion US dollars in 2025, an all-time record.
Meta booked around $200 billion US dollars in revenue in 2025. Meanwhile, as mentioned above, Block carried out layoffs not because of losses, but because it believes a smaller team with AI assistance will be more productive.
This phenomenon has sparked concerns from the world’s most influential thinkers.
Daron Acemoglu, MIT economist and winner of the 2024 Nobel Prize in Economics, has warned: “Currently, we have the wrong direction of AI development. We are using AI too much for automation, rather than to provide expertise and information that empowers workers. Looking at the prevailing paradigm in the technology industry today, we cannot rule out the worst-case scenario of all possibilities: none of AI’s truly transformative potential is realised, while all of its negative impacts, such as workforce displacement, disinformation, and manipulation, occur fully.” (MIT Economics, 2024; Project Syndicate, 2024).
He also mentioned the concept of so-so automation, which is automation that cuts labour costs without significantly increasing productivity.