Researchers from Princeton University and the University of Chicago conducted experiments using large language models (LLMs) such as ChatGPT, Claude, and Gemini in a simulated hiring environment. The study involved 20 different job scenarios designed to observe how biases might form during the AI decision-making process.
According to MIT Technology Review, the simulation aimed to better understand the behavior of these advanced AI models in recruitment contexts, highlighting potential risks of bias embedded within their algorithms. Such findings contribute to ongoing discussions about fairness and transparency in AI-driven hiring tools.
With Japan's labor market increasingly adopting AI and automation technologies, insights into AI bias are particularly relevant for local companies seeking to ensure equitable hiring practices amid digital transformation.