AI as a tool for layoffs: what the Amazon, Intel, Microsoft, Accenture and other cases show in 2024–2025
“Time for Action” analyzed how, in 2024-2025, large companies publicly and behind the scenes are restructuring their operations around artificial intelligence, while simultaneously cutting staff, revising budgets, and redefining the price of “efficiency”. At the center of this wave is not a single loud headline, but a recurring pattern: optimization is framed as “organizational changes”, “fewer layers”, “resource reallocation”, while the real outcome is often the same fewer jobs in traditional roles and higher demands on those who remain.
The most illustrative case is Amazon, where three lines intersect at once: the push for maximum automation, communication control, and growing internal social tension. According to the internal documents described, the company aims to automate a significant share of its operations, with the logic presented as purely mathematical: saving about 30 cents per order processed. Internal guidelines advise managers to avoid wording that could trigger resistance instead of “robots” or “artificial intelligence”, they are encouraged to use softer terms like “cobots” or “advanced technologies”. This detail matters, because it shows the issue is no longer only about technology, but also about how the company tries to lower emotional resistance before decisions are felt by employees.
Alongside this runs the official narrative from leadership. Amazon President and CEO Andy Jassy outlined the strategic logic as follows: “Today, practically every corner of the company is using generative artificial intelligence to make customers’ lives better and easier. We will need fewer people doing some jobs and more people doing other kinds of jobs. It’s hard to know exactly how this will play out over time, but over the next few years we expect this to reduce our total corporate workforce.” The key phrase here is “reduce our total” the reduction is not hidden, but framed as a natural consequence of “role redistribution” and training.
At the same time, the material shows that layoffs at Amazon are not an abstract forecast. It mentions Morgan Stanleyanalysts’ estimate that the company planned to cut around 14,000 managerial positions, potentially saving $3 billion annually. Then comes another internal management signal: Amazon’s Senior Vice President of People Experience and Technology Beth Galetti directly linked layoffs to the company’s shift in focus: “The generative AI revolution is the most transformational technology we’ve seen since the internet. It enables companies to innovate much faster than ever before. We believe we need to be organized more efficiently, with fewer layers and more ownership, to move as quickly as possible for our customers and business.” In corporate language, phrases like “fewer layers” and “more ownership”usually mean one thing: the structure is being compressed, and workloads are being concentrated.
The culmination is employee reaction. More than 1,000 Amazon workers signed an open letter to Jassy demanding responsible AI implementation. Their position is not about fear of innovation, but about a set of risks: climate commitments, workers’ rights, concentration of power, and the use of AI in sensitive areas. The tone of this episode shows that when a company simultaneously promotes AI as the future and speaks openly about a future with fewer people, trust erodes faster than any PR language can repair it. It is telling that one commentator summarized this as a “concentration of all these concerns, requests for structure and voice in one place”.
Against this background, other cases broaden the picture: layoffs are not limited to tech giants and do not always deliver the expected results. Paramount announced a 25% workforce reduction, justifying it as a strategic shift toward automated content production and the adoption of AI tools capable of generating scripts, editing video, and composing music. This matters because it moves the issue from “logistics and warehouses” into media and creative industries, where machines were traditionally seen as assistants, not decision-makers.
But the Klarna example in the same material shows the limits of that approach. The company cut its workforce from over 5,500 to 3,400 employees, transferred the work of 700 support agents to AI, and later admitted that customer service quality declined, announcing a return to hiring people. The key logic here is not sentimental, but market-based: brand trust and customer experience can cost more than automation that, “in theory”, was supposed to deliver efficiency.
A separate block concerns Intel, where the AI strategy overlaps with financial pressure and restructuring. The material describes how in 2024 the company cut 15,000 employees, prepared deeper cuts in 2025, laid off around 24,000 employees in July, and canceled several international projects, including planned plants in Germany and Poland that were expected to create 3,000 and 2,000 jobs respectively. At the same time, losses and concrete financial indicators are presented, which are essential for sober analysis of why layoffs become a “survival tool” rather than just faith in AI: $1.9 billion in costs related to layoffs and restructuring, a $2.9 billion loss in the second quarter on $12.9 billion in revenue, weak growth in data center business and declining PC chip sales. This is not a story of “AI replacing people”, but one where AI becomes both justification and bet amid lost market positions.
A similar but differently motivated case is Microsoft. The company confirmed layoffs of up to 9,000 employees, framed as part of organizational changes, while simultaneously investing a record $80 billion in data centers for training AI models. The material highlights a key causal link: fewer long-term and risky creative projects, more AI infrastructure, studio closures and canceled developments in the gaming division. After waves of layoffs, Satya Nadella’s position is quoted: the company will hire again, but “with much higher expectations” thanks to AI. This sets a new norm: people remain, but their roles change they are expected to produce more output in the same time, often by managing tools that raise the pace.
At Accenture, layoffs are also described as a consequence of shifting demand and active AI adoption: in one quarter, headcount fell from 791,000 to 779,000, with restructuring costs estimated at $865 million. The company plans to train 70,000 people in AI technologies, but openly acknowledges the limit of reskilling: where it does not work, people will be laid off. This is a harsher but more honest signal: reskilling becomes not a guarantee, but a filter.
The final note in the material is the “opinions” of chatbots, interesting not as machine psychology, but as a reflection of social tension. The strongest statement places responsibility where it belongs: “I’m not taking jobs companies are. I’m a tool. Where people were forced for years to work like an algorithm, the algorithm replaces them. And where work remains genuinely human with complexity, doubt, responsibility, and meaning I can at best be an assistant, not a replacement.” What matters here is not self-justification, but the idea that the mechanization of labor began long before AI, and AI simply completed it in processes where people had already been treated as parts of a conveyor belt.
AI in corporate strategies became not a standalone technology, but a universal argument for shifting the balance between labor and capital. Companies simultaneously invest in infrastructure and cut human costs, explaining it through efficiency, speed, and “fewer layers”. But the real effect has three layers. The first is structural layoffs in roles that are easy to standardize or break into repetitive tasks. The second is higher demands on those who remain: more responsibility, more pace, more functions per role. The third is the social and reputational cost, which businesses begin to feel when automation undermines service quality or triggers internal resistance, as in Amazon’s case. Against this backdrop, the main question is no longer whether “robots will replace people”, but how companies will distribute the gains from automation, who will bear responsibility for the consequences, and whether they can prove that their technological choices do not turn people into expendable resources in the pursuit of savings and control.
About the Author
Related posts:
How Artificial Intelligence Is Changing the Labor Market: Analytics on Layoffs, New Professions and Challenges in 2025 
Artificial Intelligence and the Labor Market: Which Jobs Will Disappear, Which Will Emerge, and How to Adapt 
Artificial Intelligence and Science: Has AI Come Closer to Nobel-Level Discoveries? 
OpenAI Moves Into Health: How Artificial Intelligence Is Changing Digital Medicine
