Heuristics and Biases in AI Decision-Making: Implications for Responsible AGI

TL;DR

This study evaluates cognitive biases in three large language models, revealing varying levels of bias and inconsistency, and highlights the need for improved reasoning and bias mitigation in developing responsible AGI.

Contribution

It provides a comprehensive experimental analysis of biases in LLMs, highlighting their strengths and weaknesses, and underscores the importance of addressing biases for responsible AGI development.

Findings

GPT-4o performed best overall

Gemma 2 effectively addressed certain biases

Llama 3.1 showed frequent inconsistencies

Abstract

We investigate the presence of cognitive biases in three large language models (LLMs): GPT-4o, Gemma 2, and Llama 3.1. The study uses 1,500 experiments across nine established cognitive biases to evaluate the models' responses and consistency. GPT-4o demonstrated the strongest overall performance. Gemma 2 showed strengths in addressing the sunk cost fallacy and prospect theory, however its performance varied across different biases. Llama 3.1 consistently underperformed, relying on heuristics and exhibiting frequent inconsistencies and contradictions. The findings highlight the challenges of achieving robust and generalizable reasoning in LLMs, and underscore the need for further development to mitigate biases in artificial general intelligence (AGI). The study emphasizes the importance of integrating statistical reasoning and ethical considerations in future AI development.