# Personalized Constitutionally-Aligned Agentic Superego: Secure AI Behavior Aligned to Diverse Human Values

arXiv: https://arxiv.org/abs/2506.13774 · HTML: https://mlsanity.com/paper/2506.13774/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 25 of its 50 references that are in the map.
- First posted on arXiv 8 Jun 2025.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)

## Builds on (most cited first, 25 of 25)

- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2307.15217/) (2023)
- [Artificial Intelligence, Values and Alignment](https://mlsanity.com/paper/2001.09768/) (2020)
- [AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents](https://mlsanity.com/paper/2410.09024/) (2024)
- [Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation](https://mlsanity.com/paper/2503.11926/) (2025)
- [AI Control: Improving Safety Despite Intentional Subversion](https://mlsanity.com/paper/2312.06942/) (2023)
- [Do the Rewards Justify the Means? Measuring Trade-Offs Between Rewards and Ethical Behavior in the MACHIAVELLI Benchmark](https://mlsanity.com/paper/2304.03279/) (2023)
- [Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming](https://mlsanity.com/paper/2501.18837/) (2025)
- [Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs](https://mlsanity.com/paper/2502.17424/) (2025)
- [Who Validates the Validators? Aligning LLM-Assisted Evaluation of LLM Outputs with Human Preferences](https://mlsanity.com/paper/2404.12272/) (2024)
- [OASIS: Open Agent Social Interaction Simulations with One Million Agents](https://mlsanity.com/paper/2411.11581/) (2024)
- [InferAligner: Inference-Time Alignment for Harmlessness through Cross-Model Guidance](https://mlsanity.com/paper/2401.11206/) (2024)
- [Project Sid: Many-agent simulations toward AI civilization](https://mlsanity.com/paper/2411.00114/) (2024)
- [Personalized Large Language Models](https://mlsanity.com/paper/2402.09269/) (2024)
- [Confabulation: The Surprising Value of Large Language Model Hallucinations](https://mlsanity.com/paper/2406.04175/) (2024)
- [Dynamic Search for Inference-Time Alignment in Diffusion Models](https://mlsanity.com/paper/2503.02039/) (2025)
- [Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements](https://mlsanity.com/paper/2410.08968/) (2024)
- [ShieldGemma 2: Robust and Tractable Image Content Moderation](https://mlsanity.com/paper/2504.01081/) (2025)
- [Collaborating Action by Action: A Multi-agent LLM Framework for Embodied Reasoning](https://mlsanity.com/paper/2504.17950/) (2025)
- [On Almost Surely Safe Alignment of Large Language Models at Inference-Time](https://mlsanity.com/paper/2502.01208/) (2025)
- [Disentangled Safety Adapters Enable Efficient Guardrails and Flexible Inference-Time Alignment](https://mlsanity.com/paper/2506.00166/) (2025)
- [Revealing economic facts: LLMs know more than they say](https://mlsanity.com/paper/2505.08662/) (2025)

## Built on by (most cited first, 1 of 1)

- [PASTABench: Proactive Assessment of Sequential Trajectories for Agent Safety](https://mlsanity.com/paper/2609.28197/) (2026)

## Abstract (arXiv)

Agentic AI systems, possessing capabilities for autonomous planning and action, show great potential across diverse domains. However, their practical deployment is hindered by challenges in aligning their behavior with varied human values, complex safety requirements, and specific compliance needs. Existing alignment methodologies often falter when faced with the complex task of providing personalized context without inducing confabulation or operational inefficiencies. This paper introduces a novel solution: a 'superego' agent, designed as a personalized oversight mechanism for agentic AI. This system dynamically steers AI planning by referencing user-selected 'Creed Constitutions' encapsulating diverse rule sets -- with adjustable adherence levels to fit non-negotiable values. A real-time compliance enforcer validates plans against these constitutions and a universal ethical floor before execution. We present a functional system, including a demonstration interface with a prototypical constitution-sharing portal, and successful integration with third-party models via the Model Context Protocol (MCP). Comprehensive benchmark evaluations (HarmBench, AgentHarm) demonstrate that our Superego agent dramatically reduces harmful outputs -- achieving up to a 98.3% harm score reduction and near-perfect refusal rates (e.g., 100% with Claude Sonnet 4 on AgentHarm's harmful set) for leading LLMs like Gemini 2.5 Flash and GPT-4o. This approach substantially simplifies personalized AI alignment, rendering agentic systems more reliably attuned to individual and cultural contexts, while also enabling substantial safety improvements. An overview on this research with examples is available at https://superego.creed.space.