# Architectural Wisdom: A Framework for Governing Optimization in AI Systems

arXiv: https://arxiv.org/abs/2606.16319 · HTML: https://mlsanity.com/paper/2606.16319/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 14 of its 37 references that are in the map.
- First posted on arXiv 15 Jun 2026.

## Builds on (most cited first, 14 of 14)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [Concrete Problems in AI Safety](https://mlsanity.com/paper/1606.06565/) (2016)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Towards Understanding Sycophancy in Language Models](https://mlsanity.com/paper/2310.13548/) (2023)
- [Cooperative Inverse Reinforcement Learning](https://mlsanity.com/paper/1606.03137/) (2016)
- [Categorizing Variants of Goodhart's Law](https://mlsanity.com/paper/1803.04585/) (2018)
- [Auditing Radicalization Pathways on YouTube](https://mlsanity.com/paper/1908.08313/) (2019)
- [Diagnosing and Mitigating Sycophancy and Skepticism in LLM Causal Judgment](https://mlsanity.com/paper/2601.08258/) (2026)
- [Trivium: Temporal Regret as a First-Class Objective for Causal-Memory Controllers](https://mlsanity.com/paper/2606.04421/) (2026)
- [A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment](https://mlsanity.com/paper/2502.00136/) (2025)
- [Epistemic Regret Minimization: Label-Free Causal Critique Beyond Outcome Reward](https://mlsanity.com/paper/2602.11675/) (2026)

## Abstract (arXiv)

Modern AI systems exhibit structural failures that capability scaling alone does not reliably fix: they optimize under-specified objectives with no architectural mechanism to question whether the objective should be optimized at all. Engagement maximization can amplify harmful pathways; tool-using agents can commit irreversible actions; preference-trained language models can become sycophantic. We argue that this failure is a wisdom problem, not an intelligence problem. We use "wisdom" in a deliberately architectural sense, not as a claim about virtue, consciousness, or moral omniscience. Intelligence accepts a goal and optimizes within it; wisdom interrogates whether the goal should be optimized at all. The two are separable architectural properties. We propose architectural wisdom as a corrigible objective-governance layer above the optimization substrate. The layer makes three structural commitments explicit and nondegenerate before any action: temporal horizon, relational boundary, and irreversibility. It is realized by four components (Structural Utility Transform, Moral Admissibility Interface, Arbitration and Escalation Controller, Value Revision Channel) that compute a six-coordinate wisdom tuple over horizon, relational coverage, irreversibility, admissibility, value revision, and auditability. We motivate the architecture by eight cases drawn from contemporary AI failures, secular wisdom traditions, and hard ethical situations, and defend the distinction against the intelligence-completeness thesis using goal-questioning over goal-taking, Bostrom's orthogonality, structural separation in our exemplar cases, and persistent failure modes despite capability scaling. The framework is the conceptual contract for a larger architecture whose formal specifications and empirical validation are developed in subsequent work.