Numerous media reports indicate that, since the beginning of 2026, the global rate of lay-offs in the science and technology industry has exceeded last year ' s overall level. Industry tracking data show that in just over a month in early 2026, global science and technology staff reductions have been recorded at more than 30,700; if this pace is maintained, the total for the year will exceed 2025. Data from the industry tracking platform show that in 2025, some 122,000 people globally were laid off in the science and technology sector, while by September 2026 the number was close to 130,000.
In terms of global distribution, the United States is an absolutely disaster-stricken area of the current round of downsizing, with some 24,600 jobs, or more than four fifths of the global total. The Netherlands followed by some 1,700 people, mainly in the reorganization of the semiconductor equipment manufacturer ASML. India has also suffered a severe impact — a total of 12,000 to 15,000 jobs in India in the past 12 months in Oracle, PayPal, Amazon, Adidas and Meta, most severely in Indian IT centres such as Bangalore, Hyderabad, Pune and Chennai. India ' s indigenous IT enterprises have also been downsized, with the top five IT companies in the country having combined lay-offs of nearly 7,000 in 2026 and slowing the recruitment pace for their current graduates.
On the mainland of the United States, there is a pattern of lay-offs that are "large, small and medium-sized company-led". The Amazon is the largest cut-off enterprise in the current round. PayPal announced an immediate reduction of 251 staff from its San Francisco headquarters and plans to reduce the staff by an additional 20 per cent in the coming years in order to invest more money in artificial intelligence. Autodesk and Salesforce each disclosed plans for layoffs of about 1,000 persons. Industry statistics show that as of the week of 16 September, at least 659 United States employees in science and technology had been laid off or were scheduled to be laid off. Looking back at the fact that in 2024 there were more than 15,000 layoffs in Intel, more than 14,000 in Tesla and more than 10,000 in Cisco, the main sources of layoffs in the United States science and technology industry at the time.
Focusing on the west coast of the United States, more than 1,000 technology practitioners were laid off in the Bay Area in September 2026 alone. Similarly, the downsizing of smaller but growing science and technology companies in Seattle cannot be ignored — the loss of jobs scattered among small and medium-sized technology companies is superseding into unminus mass.
The core driving force of the current round of scientific and technological layoffs is not the decline in the business base, but the strategic redistribution of resources. According to several media reports, businesses are transferring capital originally spent on human resources to artificial intelligence infrastructure and research and development. It is noteworthy that the downsizing of some companies took place in a context of strong performance — ASML was still restructuring management and technical positions while reporting record sales and strong demand. This model suggests that layoffs are more the result of enterprise initiatives to build a streamlined organizational structure centred on automation and AI-driven productivity than a reactive response to financial distress. PayPal has also made it clear that the reduction in manpower was intended to "invest more artificial intelligence".
In contrast to the large-scale lay-offs in the science and technology industry, the semiconductor industry is facing a severe shortage of manpower. An analysis by McKinsey for the SEMI Foundation suggests that the United States may face a shortfall of up to 157,000 semiconductor workers by 2030. Only 3 per cent of American engineering graduates chose to enter the chip industry, and 73 per cent reported serious difficulties in filling engineering positions. Jon Taylor, Executive Vice President of the Tristar Semiconductor Sector, publicly expressed concern about the talent pipeline. Pay levels in the semiconductor sector are much higher than in the salaried science and technology jobs — starting in six digits, with experienced people earning more than $200,000 a year. However, this has not changed the brain flow. The AI heat surge has boosted the huge demand for storage and memory chips, and Washington’s policy of promoting the return of semiconductor manufacturing is accelerating the construction of plants, but the pace of talent development is far from keeping pace.
For the American people, the effects of the current round of downsizing have a clear geographical and sectoral character. The Bay and Seattle regions, where Chinese science and technology are concentrated, are the most directly affected. Among the cut-off Chinese practitioners, H-1B visa holders face a particularly pressing identity conversion window period — normally within 60 days to find a new employer or change their identity or leave the United States.
Faced with this situation, there are several directions that deserve attention. First, the talent gap in the semiconductor sector offers the possibility of career transformation. Not only is the area well paid, but the policy of the Government of the United States continues to be profitable and the related skills training and degree programmes are increasing. Second, global capacity centres and manufacturing are creating new jobs. Thirdly, for those who are still in the United States of America's science and technology industry, it is recommended to focus on the direction of the AI input of the company in which they work. – While companies that are investing in AI on a large scale are still downsizing, their AI-related jobs may be expanding.
In the specific response steps, it is recommended that the laid-off staff immediately organize the progress of the immigration application materials I-140 and I-485, assess whether they have been approved with a labour card or are in a state of priority; update the curriculum vitae to highlight skills labels related to AI, automation; and actively contact recruitment channels in areas where talent is scarce, such as semiconductors, advanced manufacturing, etc. The window for career conversion is limited and it is recommended that a substantial job search or training programme be initiated within 30 days of the dismissal.