According to MIT Technology Review, artificial intelligence systems are increasingly being deployed to screen job applications before human recruiters review them, but new research raises concerns about the fairness of these AI-powered hiring tools.
Researchers have found that large language models (LLMs) not only inherit human biases present in their training data, but can also develop their own biases when evaluating candidates. The MIT Technology Review report indicates that AI is more likely than humans to form biases during the hiring process, though the article excerpt does not provide specific details about the nature of these additional biases or the methodology used in the research.
The findings add to growing concerns about the use of AI in employment decisions, as companies continue to adopt these technologies to manage large volumes of job applications. While AI screening tools are marketed as efficiency improvements, this research suggests they may introduce fairness issues that extend beyond simply replicating existing human prejudices.