The Same Email, Signed Differently: Testing Negotiation Bias and Recommendation Stability in LLMs

Jasmin Heierli, Alexandre de Spindler

Track: Scientific Track

Keywords: Bias, Decision Stability, LLMs, Human-AI Interaction

TL;DR: Using LLMs on both sides of hiring introduces provider-dependent and unstable decisions, even when evaluation scores appear consistent.

Abstract: Large language models (LLMs) are increasingly involved in hiring communication, both as tools that help applicants draft negotiation emails and as systems used to evaluate them. Such mediation risks introducing variability and hidden dependencies into high-stakes outcomes like salary expectations and hiring decisions. We study this bidirectional setting with a two-stage analysis across providers and English/German contexts, using 2,880 Stage 1 observations and 1,441 paired Stage 2 evaluations. We find no strong or consistent pooled gender effects. Instead, provider differences dominate, while scalar ratings are stable on average and categorical recommendations are less robust.

Submission Number: 46