# Philosophical Dispositions as Behavioral Constraints for AI-Assisted Code Review: An Empirical Study

arXiv: https://arxiv.org/abs/2605.23108 · HTML: https://mlsanity.com/paper/2605.23108/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 21 references that are in the map.
- First posted on arXiv 21 May 2026.

## Builds on (most cited first, 6 of 6)

- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation](https://mlsanity.com/paper/2308.08155/) (2023)
- [MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework](https://mlsanity.com/paper/2308.00352/) (2023)
- [Discovering Language Model Behaviors with Model-Written Evaluations](https://mlsanity.com/paper/2212.09251/) (2022)
- [LLM Critics Help Catch LLM Bugs](https://mlsanity.com/paper/2407.00215/) (2024)
- [Pramana: Fine-Tuning Large Language Models for Epistemic Reasoning through Navya-Nyaya](https://mlsanity.com/paper/2604.04937/) (2026)

## Abstract (arXiv)

AI-assisted code review tools typically operate as generic "expert reviewer" agents, producing homogeneous findings regardless of the analysis type needed. We present a system that constrains AI reviewer behavior through philosophical dispositions -- coherent personality lenses grounded in specific epistemological traditions (Pyrrhonist Skepticism, Navya-Ny=aya logic, Diogenes' Cynicism, Confucian relational ethics) that direct attention to structurally different types of issues. Each disposition is defined apophatically (by what it refuses to do), equipped with a self-monitoring failure mode (hamartia), and orchestrated in sequence by role protocols. We evaluate this system on 50 merged pull requests across 7 repositories spanning 5 programming languages (Python, Go, C++, Java, Terraform), 5 organizations (2 enterprise, 3 open-source), and 2 temporal eras (pre-AI 2020, post-AI 2024--2026). The disposition system achieves 46% convergence with human reviewers (validating signal quality), identifies unique findings at a 75% rate, and produces no findings judged false-positive by the author across 601 total findings (inter-rater agreement was not assessed and remains a limitation). A controlled baseline comparison demonstrates that 51% of disposition findings are not produced by the same model using generic "expert reviewer" prompting, and these unique findings target structural, operational, and logical concerns rather than standard code-level issues. Preliminary cross-model validation (Claude Opus vs. GPT Codex 5.3-xhigh) on 3 PRs shows 100% framework-structure adherence with 39% finding-level agreement, suggesting the framework provides real behavioral constraint while preserving model-specific analytical perspective.