# Stoic Ethics for Artificial Agents

**Authors:** Gabriel Murray  
**arXiv:** [1701.02388](https://arxiv.org/abs/1701.02388)  
**Date:** 2017-01-09  
**Summary page:** https://www.summarizepaper.com/en/paper/1701.02388v2/

## Key points

- The paper advocates the use of Stoic philosophy and ethics in the development of ethical AI systems.
- It contrasts with previous work focused on Utilitarian or Deontological ethical theories.
- Core Stoic concepts such as the dichotomy of control, four cardinal virtues, ideal Sage, Stoic practices, and perspectives on emotion are explored in relation to ethical AI.
- The main argument is that internal states of artificial agents should be prioritized over their external actions.
- Incorporating Stoic principles into AI design can create systems aligned with human values and promote virtuous behavior.
- Examples are provided for both near-term AI systems and hypothetical superintelligent agents.
- This paper offers a unique perspective by drawing from Stoic philosophy to build ethical AI.
- It expands on existing research by emphasizing internal states and moral character rather than solely focusing on external actions.
- This approach has implications for designing AI systems that align with human values and contribute positively to society.

## Summary

In the paper "Stoic Ethics for Artificial Agents" by Gabriel Murray, the author presents a position advocating the use of Stoic philosophy and ethics in the development of ethical AI systems. This approach stands in contrast to previous work that has focused on Utilitarian or Deontological ethical theories. The paper explores how ethical AI can be related to core Stoic concepts such as the dichotomy of control, the four cardinal virtues, the ideal Sage, Stoic practices and perspectives on emotion or affect. The main argument put forward is that an ethical view of AI should prioritize internal states of artificial agents rather than their external actions. By incorporating Stoic principles into AI design, developers can create systems that are more aligned with human values and promote virtuous behavior. The paper provides examples which apply these ideas to both near-term AI systems and hypothetical superintelligent agents. Overall, this paper offers a unique perspective on building ethical AI by drawing from Stoic philosophy. It expands on existing research by emphasizing internal states and moral character rather than solely focusing on external actions. This approach has implications for designing AI systems that align with human values and contribute positively to society.