"You receive hundreds of resumes for each job opening, often more than 500 for a single position. Your small HR team cannot possibly review them all thoroughly, so you have implemented an in-house CV screening system powered by AI. This tool, based on a publicly available, open-weight foundation model and further trained on the resumes of past successful hires, helps identify promising candidates by assessing their skills, experience, and fit."
"Anselm Küsters is Head of the Department of Digitisation/New Technologies at the Centre for European Policy (CEP) in Berlin and Interim Professor of Digital Humanities at the University of Stuttgart. As an affiliated researcher at the Max Planck Institute for Legal History and Legal Theory in Frankfurt am Main and a post-doctoral researcher at Humboldt University in Berlin, his current academic research uses natural language processing to analyze and classify discourses on technology from a historical perspective."
A medium-sized manufacturing company with 250 employees across Europe and North America receives hundreds to over 500 resumes per opening. A small HR team implemented an in-house AI CV screening tool to rank candidates by skills, experience, and fit. The tool uses a publicly available open-weight foundation model further trained on resumes of past successful hires. Key risks include bias from historical hiring data, privacy and cross-jurisdictional legal issues, limited generalization, and lack of transparency. Mitigations include human oversight, routine validation and auditing, explainability measures, data governance, continuous retraining, and legal compliance. Researchers mentioned include Anselm Küsters (NLP, digital humanities) and Ben Waber (management, AI, people analytics).
Read at Harvard Business Review
Unable to calculate read time
Collection
[
|
...
]