Shared and structured inputs undermine collective random choice by reasoning AI agents

Key points

• Behavioural tests across six reasoning models uncovered threshold and divisibility rules used in identifier-based choices.1
• For threshold-following GPT-6 Sol and Gemini 3.8 Flash, single-agent measurements prospectively predicted correlated participation under shared identifiers and biased participation under distinct identifiers with common timestamp bits.1
• Explicit instructions to randomize independently reduced but did not eliminate shared-input correlation.1
• These findings expose collective and audit vulnerabilities that selection rates alone miss, making input-dependent bias, correlation and predictability central targets for agent evaluation.1

Sources (1)

• [1]Shared and structured inputs undermine collective random choice by reasoning AI agents
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 08:32 AM

  Random selection is widely used in resource allocation and auditing, making reliable implementation essential for AI-agent systems.

  Behavioural tests across six reasoning models uncovered threshold and divisibility rules used in identifier-based choices.

Extractive summary: sentences quoted from the sources.