New research reveals: specific work experience affects the effectiveness of using artificial intelligence in the workplace
New research reveals that work experience has a significant impact on how employees are more assisted by artificial intelligence. Employees with more specific task experience use AI more, while veteran employees are less likely to distrust AI due to fear of its shortcomings. The findings highlight the need for customized strategies for integrating artificial intelligence in the workplace to improve human collaboration with artificial intelligence.
New research from the Virginia Institute for Performance Research and Management Science published in the INFORMS journal Management Science offers valuable insights for business managers about the impact of work experience on employee interactions with artificial intelligence.
The study examines the effect of two main forms of human work experience - narrow experience, defined by the amount of specific tasks, and broad experience, characterized by general seniority - on the dynamics in human teams with artificial intelligence.
Surprising findings from a study that dealt with coding medical files
"We developed an artificial intelligence solution for coding medical files in a public company, and conducted a study among the company's knowledge workers," says Weiguang Wong from the University of Rochester and lead author of the study. "We were surprised by what we found in the study. The different dimensions of work experience have different interactions with artificial intelligence and unique roles in human collaboration with artificial intelligence."
"While one might think that less experienced workers would benefit more from the use of artificial intelligence, we find the opposite - artificial intelligence benefits workers with more experience in the specific task. At the same time, veteran workers, despite their greater experience, derive less benefit from artificial intelligence than their younger colleagues ", says Gordon Gao of the Carey School of Business at Johns Hopkins University, and co-author of the study.
The dilemma of seniority and trust in artificial intelligence
The continuation of the study reveals that the relatively small increase in productivity from artificial intelligence is not a result of seniority per se, but of the higher sensitivity of veteran workers to deficiencies in artificial intelligence, which reduces their trust in it.
"This finding presents a dilemma: employees with more experience are in a better position to utilize artificial intelligence to increase productivity, but the veteran employees who take on more responsibility and care for the organization tend to avoid artificial intelligence because they see the risks of relying on artificial intelligence assistance. As a result , they are not making good use of artificial intelligence," says Ritu Aggarwal of the Johns Hopkins Carey School of Business, a research partner.
The researchers urge employers to carefully consider different types and levels of employee experience when introducing artificial intelligence into the workplace. New hires with less task experience are at a disadvantage in utilizing artificial intelligence. Meanwhile, veteran workers with more organizational experience may fear the potential risks posed by artificial intelligence. Answering these unique challenges is key to successful collaboration between people and artificial intelligence.
More of the topic in Hayadan:
- Artificial intelligence can threaten some jobs, but it is more likely to become our personal assistant
- Israel adopts international principles for innovation and responsible use of artificial intelligence
- For the first time in Israel: implementation of artificial intelligence in government offices
- A document of the Knesset Research Center: Creative Artificial Intelligence: Opportunities, Risks and Regulation Part I
Comments
From my experience working with Chatgpt to help with software development, for example in the languages Java, Assember, C, Python,
It helps a lot in saving Google searches for example if I need certain algorithms or data structures.
But, as soon as I need something different from standard things that can be found on Google then sometimes
ChatGTP leads in a problematic or even completely wrong direction.
Therefore in my experience, it can mainly be used as a smart search engine.
Eli Isaac is a senior software engineer and a private computer science teacher
https://go100.co.il/private_teacher_programming.html
Something is confused in the article or is it me
The title is misleading. In the article there is no mention of an effect on the number of vacant jobs (taking the workers' jobs)
It may have been hinted at, but not only is there no support for it, it's not even the research question here