AI Research
Fair outputs, Biased Internals: Causal Potency and Asymmetry of Latent Bias in LLMs for High-Stakes Decisions
Instruction-tuned language models exhibit behavioural fairness in high-stakes decisions while retaining biased associations in their internal representations. However, whether these suppressed representations can affect model outputs - and whether such causal potency is symmetric...
AC
ArXiv CS.AI
Source
2 min
ArXiv CS.AIRead the full story at ArXiv CS.AI
AI Analysis
Generated insight by OnAirToday AI
This research highlights a critical disconnect: while LLMs may appear fair in their outputs, they can harbor and even amplify hidden biases in their internal data. These latent biases are not inert and can be exploited to reverse decisions, underscoring the inadequacy of output-only fairness evaluations for high-stakes AI applications.
Powered by AI • Not a substitute for human analysisMore Insights →
Share
Filed under:AI Research
Comments
Keep comments relevant and respectful. No links, advertising, self-promotion, or spam.