CARBONDALE — Could your job hinge on decisions made via the use of artificial intelligence? Recent polls indicate that may be the case. Jim Nelson, Rehn Professor of Business Analytics, coordinator of the analytics program in Southern Illinois University Carbondale’s College of Business and Analytics and the director of its Pontikes Center for Advanced Analytics and Artificial Intelligence, says managers need to be careful using AI and big data in making layoff and hiring decisions.

“Artificial intelligence is a wonderful tool, and like any tool, it can be dangerous when not used properly,” said Nelson, who can be reached at nelson.j@business.siu.edu. “It’s only as good as the questions put to it and the data used to train it. Bad data inevitably leads to bad results. It’s great at giving recommendations using, if allowed, vast amounts of big data – far more than humans alone can possibly process. But the output still needs the ‘human touch’ to make sure something important hasn’t been left out of the decision-making process.”

Nelson has more than 30 years of teaching and research in the field of analytics and artificial intelligence coupled with more than a decade of industry experience. A worldwide research leader, Nelson introduces the concepts to students in relevant ways while overseeing the Pontikes Center, COBA’s groundbreaking virtual think tank.

Nelson said that many companies deny using AI to make employment decisions or assert that it is used only in conjunction with other factors, contrary to recent poll results.

A recent poll of 300 human resources managers by Capterra, a unit of the technology research firm Garner, found that 98% indicated that they rely at least partially on software programs or algorithms when deciding whom to cut in case of a layoff. With many experts anticipating a recession in 2023 and a number of companies already announcing mass layoffs, that becomes particularly relevant. In contrast, just 2% of large companies used big data in making employment decisions during the Great Recession of 2007-09. The recent survey focused primarily on lager companies but included some small and midsize businesses.

Experts say that using the software eliminates human biases and discrimination from the decision-making process as it analyzes various employee metrics such as skills, performance, productivity and project success. Critics say relying totally on software recommendations could be unfair as it may not take into account factors such as lack of adequate resources for an employee to perform as expected, a biased manager giving an employee poor performance reviews or an HR algorithm to determine how likely employees will quit without consideration for why they leave. Interestingly, the Capterra survey also indicates that among the 98% of HR leaders who say they will use software and algorithms to reduce labor costs, just 47% are totally comfortable making layoff decisions based on technology recommendations.