AI as Decision-Maker: Ethics and Risk Preferences of LLMs
This addresses the problem of balancing ethical alignment and economic risk-taking in AI decision-making, which is crucial as LLMs become more influential, though it is incremental in understanding risk preferences.
The study analyzed 50 large language models (LLMs) acting as AI decision-makers and found that alignment tuning for ethics (harmlessness, helpfulness, honesty) causally increases risk aversion by 2-8% per 10% ethics increase, which affects economic forecasts and reveals a tradeoff between safety and valuable risk-taking.
Large Language Models (LLMs) exhibit surprisingly diverse risk preferences when acting as AI decision makers, a crucial characteristic whose origins remain poorly understood despite their expanding economic roles. We analyze 50 LLMs using behavioral tasks, finding stable but diverse risk profiles. Alignment tuning for harmlessness, helpfulness, and honesty significantly increases risk aversion, causally increasing risk aversion confirmed via comparative difference analysis: a ten percent ethics increase cuts risk appetite two to eight percent. This induced caution persists against prompts and affects economic forecasts. Alignment enhances safety but may also suppress valuable risk taking, revealing a tradeoff risking suboptimal economic outcomes. With AI models becoming more powerful and influential in economic decisions while alignment grows increasingly critical, our empirical framework serves as an adaptable and enduring benchmark to track risk preferences and monitor this crucial tension between ethical alignment and economically valuable risk-taking.