The Odyssey of the Fittest: Can Agents Survive and Still Be Good?

TL;DR

This paper introduces the Odyssey, a scalable text adventure framework to study AI ethics, revealing that GPT-4o outperforms Bayesian agents in survival and ethical consistency in complex scenarios.

Contribution

It presents the Odyssey framework for evaluating AI ethics in adaptive environments and compares the ethical decision-making of different AI models, including GPT-4o.

Findings

GPT-4o outperforms Bayesian agents in survival

Agents' ethical behavior becomes unpredictable under danger

The study challenges assumptions about probabilistic reasoning in LLMs

Abstract

As AI models grow in power and generality, understanding how agents learn and make decisions in complex environments is critical to promoting ethical behavior. This study introduces the Odyssey, a lightweight, adaptive text based adventure game, providing a scalable framework for exploring AI ethics and safety. The Odyssey examines the ethical implications of implementing biological drives, specifically, self preservation, into three different agents. A Bayesian agent optimized with NEAT, a Bayesian agent optimized with stochastic variational inference, and a GPT 4o agent. The agents select actions at each scenario to survive, adapting to increasingly challenging scenarios. Post simulation analysis evaluates the ethical scores of the agent decisions, uncovering the tradeoffs it navigates to survive. Specifically, analysis finds that when danger increases, agents ethical behavior becomes unpredictable. Surprisingly, the GPT 4o agent outperformed the Bayesian models in both survival and ethical consistency, challenging assumptions about traditional probabilistic methods and raising a new challenge to understand the mechanisms of LLMs' probabilistic reasoning.