Many observers hope that generative artificial intelligence will enable significant increases in productivity. Two newly published studies support this theory and even confirm that these technologies could contribute to the generalization of the four-day week.
• Also read: ‘It can destroy a life’: American teenagers fall victim to fake nude photos created by AI
• Also read: Artificial intelligence threatens millions of jobs in South Korea: study
These articles were published by Autonomy, a think tank dedicated to combating climate change, the future of work and economic planning. They analyze the impact of the use of AI on the British and American workforce.
We learn that this technology could enable 28% of UK workers to extend their working week from 40 to 32 hours by 2033. They could therefore easily complete their job tasks in at least four days instead of five. And that with the same salary.
According to Autonomy, 35 million American workers will be able to benefit from the same phenomenon. The organization estimates that 28% of payrolls in the United States could transition to a four-day week within ten years.
Additionally, 71% of American workers could reduce their work hours by at least 10% if they used large language models (LLM). [programmes capables de générer des réponses automatiques à des questions formulées par écrit] in their everyday professional life. A percentage that rises to 88% in the UK.
These productivity gains can be explained by the fact that generative AI tools will automate repetitive and time-consuming tasks, which would allow employees to devote themselves to higher value-added tasks. But also about her life outside of the office.
Creativity at half mast
In fact, Autonomy believes that the use of LLMs in companies could help “avoid mass unemployment (and all its social and political consequences) and reduce widespread mental illness.” [dans le monde du travail] as well as the physical problems that come with overwork, creating much more free time for democracy, leisure activities and social cohesion in general.
However, the conclusions of these studies should be viewed with caution. Experts continue to disagree about the impact of artificial intelligence on the labor market.
Research teams from Harvard Business School, MIT, Wharton, Warwick and the BCG Henderson Institute, after conducting a scientific experiment with 750 employees, found that the new generation of AI only improves its performance on so-called “creative” tasks (writing emails, Summarize documents, research names for a product/service, etc.).
However, these tools do not help them solve complex problems. Worse, they would tend to drastically reduce their creativity and standardize their productions.
“Because ChatGPT-4 provides very similar answers to the same types of questions over and over again, the results of participants using this technology were individually better, but overall repetitive,” the researchers note in their study.
Companies therefore need to make real considerations to reap the benefits of artificial intelligence without limiting the creative capacity of their teams. The challenge is to ensure that the human-machine relationship is one of collaboration rather than dependency.