AI as Decision-Maker: Ethics and Risk Preferences of LLMs
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908400961454080 |
|---|---|
| author | Ouyang, Shumiao Yun, Hayong Zheng, Xingjian |
| author_facet | Ouyang, Shumiao Yun, Hayong Zheng, Xingjian |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_01168 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | AI as Decision-Maker: Ethics and Risk Preferences of LLMs Ouyang, Shumiao Yun, Hayong Zheng, Xingjian General Economics Economics Artificial Intelligence Computers and Society Emerging Technologies Human-Computer Interaction 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. |
| title | AI as Decision-Maker: Ethics and Risk Preferences of LLMs |
| topic | General Economics Economics Artificial Intelligence Computers and Society Emerging Technologies Human-Computer Interaction |
| url | https://arxiv.org/abs/2406.01168 |