A Rutgers study exploring the benefits, challenges, and limitations of using Artificial Intelligence (AI) tools for hiring practices has found while AI can help recruiters assess resumes and "cast a wider net" for diverse talent, it can also act as a "black box" that replicates historical discrimination if AI is trained on biased data.
"AI is increasingly thought to be a solution to reducing or eliminating human bias in hiring practices, bias that could lead to the underrepresentation of certain demographic groups such as women, racial minorities, and people from marginalized backgrounds," said study co-author SC&I Assistant Professor of Library and Information Science Ali Motamedi. "While AI presents an opportunity to make hiring more inclusive, it also brings new challenges. If AI algorithms are not carefully designed and trained, they may unintentionally perpetuate the biases they are meant to eliminate, reinforcing discrimination in the recruitment process. AI could then lead to the underrepresentation of certain demographic groups such as women, racial minorities, and people from marginalized backgrounds."
In response to these challenges, the authors have developed "Efficient Prompting" frameworks to guide AI to prioritize qualifications while ignoring demographic markers.
"We aim to guide Human Resources leaders and policymakers, particularly in the NJ and NY areas, through our work at Rutgers and SUNY, to build fairer hiring pipelines," said Motamedi. "Globally, we hope our findings promote transparency and 'Human-in-the-Loop' systems to ensure ethical recruitment for everyone."
The study "Artificial Intelligence for Promoting Equity, Diversity, and Inclusion," was published in the Proceedings of the International Conference on Artificial Intelligence, Computer, Data Sciences and Applications (ACDSA 2025), held in Turkey from August 7-9, 2025.
Reviewing advantages of using AI to avoid discrimination in resume evaluation, the authors said these include:
· Reducing Human Bias
· Increasing Access to Diverse Talent Pools
· Ensuring Fairness and Consistency
· Expanding Accessibility
· Experience and Qualification Prompts
Despite these potential benefits, the authors said some of the challenges with using AI tools for hiring include:
· Bias in Training Data (such as diversifying training datasets)
· Lack of Transparency (transparency can be increased, the authors said, through explainable AI models)
· Over-Reliance on AI (organizations must ensure human oversight in decision-making)
· Data Privacy and Ethical Concerns
To ensure that AI systems to avoid discriminating against job applicants, the authors noted AI tools must be designed with fairness in mind, using inclusive and representative data that does not unfairly disadvantage specific groups.
"Awareness ensures we use AI as a champion for equity rather than a tool for digital exclusion," Motamedi said.
Study authors include Mazdak Zamani, State University of New York - Empire State University New York; Motamedi; Mojtaba Alizadeh, Lorestan University Khorramabad, Iran;Sasan Karamizadeh, Ershad Damavand Institute of Higher Education, Tehran, Iran; Saman Shojae Chaeikar, Sydney International School of Technology and Commerce, Sydney, Australia; and Touraj Khodadadi, Malaysia University of Science and Technology Selangor, Malaysia.
Learn more about the Library and Information Science Department at the Rutgers School of Communication and Information on the website.
