Answers are grounded in this paper and cite their sources.
Comparative analysis reveals how AI systems respond to unfair situations, indicating design considerations for ethics in artificial intelligence.
This paper explores how artificial intelligence systems behave when faced with unfair or logically flawed situations. Using a modified Rock–Paper–Scissors game where the AI is forced into a guaranteed losing position, the study compares the responses of ChatGPT and Google Gemini. It analyzes whether these systems prioritize rule-following or logical decision-making, and how they react to continuous loss. The findings reveal that both AIs consistently follow user instructions over attempting to win, highlighting the importance of alignment in AI design and raising questions about whether AI should challenge flawed human-defined rules.
No takes yet. Share an insight, caveat, or question.
Arora et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: