AI-driven hiring platforms have become increasingly popular among businesses as they aim to streamline the recruitment process and eliminate biases. However, concerns are growing as highly qualified candidates are being overlooked by these algorithms. Despite the initial hope that AI recruiting technology would bring fairness to the hiring process, it seems that the opposite is happening.
Hilke Schellmann, an assistant professor of journalism at New York University and author of “The Algorithm: How AI Can Hijack Your Career and Steal Your Future,” believes that these tools are not accurately screening the most qualified candidates. In fact, she argues that the biggest risk posed by this software is not machines taking over jobs, but rather preventing qualified individuals from getting hired at all.
There have been cases where highly qualified job candidates have found themselves at odds with these hiring platforms. Anthea Mairoudhiou, a UK-based make-up artist, shared her experience of being evaluated by an AI-screening program called HireVue. Despite ranking well in the skills evaluation, she was ultimately rejected due to a poor score in body language analysis. Similar complaints have been filed against other platforms, highlighting the potential flaws in their algorithms.
One of the main concerns is that these algorithms are often trained on a specific type of employee, which means candidates with different backgrounds or credentials may be filtered out. Schellmann points out that marginalized groups are particularly vulnerable to being overlooked by these systems due to differences in hobbies or educational backgrounds.
The selection criteria used by these algorithms can be biased and sometimes even opaque. Schellmann conducted research where she applied for a call center job and was screened by AI. Despite speaking nonsense German instead of English during the interview, she received a high rating. However, her actual relevant credentials on her LinkedIn profile received a poor rating. This highlights the potential flaws in the evaluation process.
The negative effects of biased hiring platforms are likely to spread as the technology becomes more widespread. Schellmann warns that an algorithm used by a large company to screen all incoming applications could harm hundreds of thousands of applicants. The problem lies in the fact that no one knows exactly where the harm is, and companies may have little motivation to address these issues as they have saved money by replacing human HR staff with AI.
Sandra Wachter, a professor of technology and regulation at the University of Oxford’s Internet Institute, emphasizes the importance of developing unbiased and fair AI. She believes that not only is it ethically and legally necessary, but it also makes a company more profitable. Wachter has been working on the Conditional Demographic Disparity test, a tool that helps companies identify bias in their algorithms and make adjustments to ensure fairness.
In order to prevent these issues from persisting, Schellmann calls for industry-wide regulations and guardrails from governments or non-profits. Without intervention, she fears that AI could make the workplace even more unequal in the future.
It is clear that while AI-driven hiring platforms have the potential to streamline the recruitment process, they also come with their own set of challenges. The concerns raised by experts highlight the need for careful development and regulation of these technologies to ensure fairness and equal opportunities for all candidates.