Developing story · 2 episodes · 20 Oct to 20 Oct
Measuring Bias in Artificial Intelligence Systems
The saga explores the tension between AI's potential to amplify human bias and the emerging tools designed to measure and mitigate these disparities. After initial criticism, the release of the FairStress benchmark marks a concrete step toward standardizing fairness evaluations in large language models.
-
Ai2 releases FairStress benchmark to evaluate demographic fairness in LLM decisions
Ai2 introduced FairStress, a benchmark that measures hidden demographic fairness asymmetries in large language models. It contains 300 realistic two‑candidate decision scenarios across 29 domains…
1 source primary source -
Opinion: AI scales human bias instead of eliminating it
AI systems inherit the judgments and blind spots embedded in the data they are trained on, so when they are used for hiring, healthcare or employee evaluation they can amplify existing…
1 source
Who and what