Osman Şenkul
Assistant Professor Morgan Frank from the Department of Computer and Information Systems at the University of Pittsburgh’s School of Computer and Information Sciences, who studies the unemployment risk posed by artificial intelligence across professions, found that the only workers affected by ChatGPT’s launch in November 2022 were those employed in the office and administrative support sector. Frank noted that these workers’ likelihood of receiving unemployment benefits increased in early 2023, following the release of the chatbot developed by OpenAI. However, he stated that ‘there is no significant change in the trend related to ChatGPT’s release’ for computer and mathematics occupations.
Frank said, ‘Both tech workers and administrative workers are facing a more challenging job market than a few years ago.’ He added, ‘But I don’t think AI is the reason for all of this.’ According to Frank’s findings, Amazon and many of its competitors in the technology sector rapidly hired employees in the years before the coronavirus pandemic and during the first months of the pandemic, when the Federal Reserve lowered US interest rates to near-zero levels.
Experts stated that these recruitments ultimately prepared these companies for workforce reductions – a dynamic distinct from the recent three-year surge in productive AI. The Fed began raising interest rates around the time ChatGPT was launched. Martha Gimbel, Executive Director of the Budget Lab research centre at Yale University, said in a study titled ‘Evaluating the Impact of AI on the Labour Market: Current State of Affairs’, ‘Most of the discussions on this topic sound very different to people because the word AI is involved,’ adding: ‘However, from what I have seen so far, especially at this stage of the economic cycle, there is nothing unusual in companies’ hiring and firing practices.’ Gimbel also pointed out that the question of what hiring patterns will look like when the economy enters a period of robust growth again is a big one.
Gimbel emphasised that in the long term, it will be crucial to distinguish between cyclical job losses and AI-driven job losses. According to him, if the US economy enters a recession, human resources and marketing jobs will be the expected losses; however, these roles are also exposed to artificial intelligence, making it difficult to determine whether job losses are the result of macroeconomic conditions, the adoption of technology, or both.
Nevertheless, he acknowledged that the subsequent correction following the strong employment growth after the pandemic may also have contributed to the companies’ recent layoffs.
Lawrence Schmidt, Professor of Finance at the Massachusetts Institute of Technology (MIT) Sloan School of Management, noted that Amazon, which has made the largest layoffs during this period, could automate jobs faster than most of its competitors due to its size, adding:
‘It’s not at all crazy to think that Amazon might want to eliminate certain types of roles that can be automated quickly or avoid hiring additional staff for certain types of roles. Regardless of what happens to the overall number of jobs, you can expect some reallocation.’
However, some question whether artificial intelligence is to blame and are sceptical that the recent high-profile layoffs are a clear sign of the technology’s impact on employment.
Some of America’s largest companies have begun limiting or reducing their workforce, citing promises of increased efficiency through artificial intelligence. However, so far, evidence is inconsistent as to whether the cost savings promised by artificial intelligence are truly worth the investment companies are making in this area.
This has led some experts to question whether artificial intelligence is being used as an excuse by companies laying off employees for old-fashioned reasons, such as poor financial performance or global economic uncertainty.
David Autor, Co-Director of the School of Activity and Inequality Initiative at the Massachusetts Institute of Technology and Professor of Economics, explains that it is much easier to say, ‘We are laying off employees because we are experiencing an economic slowdown, we have a bloated structure, and our profitability is low,’ than to say, ‘We are laying off employees because we are achieving productivity gains through artificial intelligence.’ we are facing an economic slowdown’.”
‘Whether artificial intelligence exists or not, it would be wise to attribute praise or blame to artificial intelligence,’ explains Autor, who is known as an expert conducting research on the impact of artificial intelligence on employees. Like other companies, Amazon, one of the world’s largest companies, announced that it had begun a restructuring that would result in the elimination of 14,000 positions, stating that artificial intelligence was the main reason for this.
Amazon Senior Vice President Beth Galetti wrote last week, ‘The world is changing rapidly.’ ‘This generation of artificial intelligence is the most transformative technology we’ve seen since the internet, and it’s enabling companies to innovate much faster than ever before.’
However, a few hours later, another Amazon representative, who did not want to be named, tried to downplay the role of artificial intelligence in the layoff decisions: “Artificial intelligence is not behind the vast majority of the layoffs. Last year, we decided to strengthen our culture and teams by, among other measures, reducing layers of hierarchy. The layoffs we announced today are a continuation of that work.”
The representative declined to comment on the apparent inconsistency between this second statement on artificial intelligence and Amazon’s previous comments. However, these statements from a large and disciplined company like Amazon highlight how difficult it can be to verify the general accuracy of what companies say about the role of artificial intelligence and personnel decisions.
According to information published in a series of news agencies and websites on the subject, Amazon has joined many companies that have recently highlighted mass layoffs, citing artificial intelligence as the reason. The US-based retail chain Walmart has also announced that it plans to keep its workforce stable over the next few years, largely due to artificial intelligence, while the US-based multinational investment bank Goldman Sachs announced a new wave of layoffs this month, stating that it plans to reduce the number of human roles that could potentially be replaced by artificial intelligence.
Salesforce, a US San Francisco-based cloud-based software company, also recently reduced its workforce by 4,000 people, citing the ‘benefits and efficiency’ of artificial intelligence.
Nitekim, yapay zeka araştırma şirketi AlphaSense’in verilerine göre, yapay zekanın yatırım getirisine odaklandığını bildiren şirketlerin sayısı son aylarda önemli ölçüde arttı. Peki, tüm bu faydalar tam olarak nerede? İşte burada işler biraz karmaşıklaşıyor. Son zamanlarda yapılan araştırmalar, en azından şu anki haliyle AI’nın üretkenliğinde önemli sınırlamalar olduğunu ortaya koydu. Boston Consulting Group’un Eylül ayında yayınladığı bir rapor için ankete katılan 1.250 şirketin yüzde 60’ı, yapay zekaya “önemli yatırımlara rağmen gelir ve maliyet açısından çok az kazanç elde ettiklerini” belirtti. Benzer bir Deloitte anketine katılan kuruluşların sadece yüzde 10’u, “ajan yapay zeka” veya basitçe komutları takip etmenin ötesinde kararlar alabilen sistemlerden “önemli bir yatırım getirisi elde ettiklerini” söyledi.
Bununla birlikte, UPenn’in Wharton School ve GBK Collective’in yeni raporuna göre, her zamankinden daha fazla büyük Amerikan şirketi, üretken yapay zekayı kullanıyor, ona yatırım yapıyor ve iş üzerindeki etkisini ölçüyor. Ancak diğer anketler gibi, Wharton raporu da karışık sonuçlar gösteriyor. Wharton’da İnsan-yapay Zeka Araştırmaları biriminden Stefano Puntoni, “Bir e-postayı 20 dakika veya bir raporu yarım saat daha kısa sürede okuyabilmek harika. Ancak bu, hiçbir şeyi atlatmanıza yardımcı olmaz” dedi.
Martha Gimbel, executive director of the Yale University Budget Laboratory, said that inferring the impact of artificial intelligence on jobs from managers’ statements during layoffs is ‘probably the worst way’ to determine its effects. Gimbel said company-specific dynamics are often more influential. ‘Because everyone is so afraid of the potential future impact of artificial intelligence on the labour market, there is a tendency to overreact to individual company statements,’ Gimbel said.
Adam Sarhan, a manager at New York-based 50 Park Investments, said, ‘Investors are asking: What does this mean? Since we can’t see the big picture, layoffs give us clues.’ He added, ‘Mass layoffs at companies like Amazon mean the economy is slowing down, not strengthening. In a strong economy, there are no mass layoffs.’
Major technology companies are laying off staff, citing advances in artificial intelligence. Experts, however, say the reality is more complex. If you read a typical 2025 mass layoff announcement from a technology sector CEO, you might think artificial intelligence is taking away workers’ jobs. The reality is more complex; companies are trying to signal that they are making themselves more efficient as they prepare for the broader changes brought about by artificial intelligence. According to a new report by job listing site Indeed, job postings in the technology sector have decreased by 36 per cent compared to 2020. However, the sole reason for this is not that companies want to replace employees with artificial intelligence. The new report states that job postings in the technology sector in July were down 36 per cent compared to early 2020 levels and that artificial intelligence is only one of the factors hindering recovery.
According to experts, there is a more prevalent factor than artificial intelligence replacing workers: companies need more money to implement artificial intelligence. The energy required to build data centres, chips and artificial intelligence systems entails significant expenditure, and technology companies are trying to justify this situation.
Bryan Hayes, a strategist at Zacks Investment Research, said that restructuring in the age of artificial intelligence is a ‘double-edged sword.’ Companies are trying to ‘find the right balance between maintaining adequate staffing levels and allowing artificial intelligence to take centre stage.’ Hayes said that larger-scale layoffs help improve profit margins, but it is difficult to say what this means for the employment prospects of the workers in question.
Consequently, a growing number of large global companies have announced that they will be directing their resources towards artificial intelligence (AI) ventures and carrying out large-scale redundancies, citing the impact of Trump’s tariff policy changes.
Although a significant proportion of these major companies announcing large-scale redundancies are headquartered in the US, their operations and, consequently, their workforce are spread globally.
Although these companies prepare and publish their quarterly and annual balance sheets at a centralised level, they disclose the balance sheets of their local companies or partnerships in the countries where they operate and prepare and announce their global and regional investment plans based on this data. These plans naturally include all personnel movements, including layoffs and hiring.
In other words, we can see more clearly that the basis of the wave of mass redundancies spreading across the globe is the reflex to ‘maximise profits’, just as it was two centuries ago. At that time, organised movements that initially emerged with machine smashing gradually reached a global level and paved the way for the International through the mass uprisings that gave birth to ‘May Day’.
Looking at it this way, the target of today’s Luddite movement should not be artificial intelligence machines, but rather, as we emphasised above, capital itself, which promotes these machines to increase profitability and cheapen labour. For this reason, as we stated in last week’s article entitled ‘Could the Luddism of two centuries ago be the Internationalism of these years?’, the path to an International uprising against today’s global wave of layoffs seems to be opening up. Because the only way to resist this global attack, which is taking away the jobs of millions of people under the pretext of artificial intelligence, is through an ‘international counter-stance’.