Radical AI Interpretability

• URL: http://arxiv.org/abs/2606.26523v1
• Date: Thu, 25 Jun 2026 01:58:38 GMT
• Title: Radical AI Interpretability
• Abstract summary: We develop a framework for interpreting AI systems as agents.<n>We draw on the philosophical tradition of radical interpretation and the tools of mechanistic interpretability.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We develop a framework for interpreting AI systems as agents, drawing on the philosophical tradition of radical interpretation and the tools of mechanistic interpretability. The core question is: given the computational facts about a system, how do we solve for its beliefs, desires, and meanings? This matters increasingly for safety. We want to be able to trust the systems we deploy, whether by understanding their goals or, more modestly, by reliably detecting deception. Interpretability researchers are building tools to read beliefs and desires off a model's internals, but there is no settled account of when such a tool has succeeded. This book supplies one. We propose criteria on both representationalist and interpretationist approaches, and tie each to tests current interpretability methods can carry out. A central lesson is that these attributions cannot be made piecemeal. Beliefs, desires, and the propositional structure they presuppose are jointly constrained, and a method that fixes one while measuring the others inherits whatever distortions that introduces. This holism becomes pressing for AI systems, which may not share the interpreter's concepts. However, it also provides leverage: a system's attitudes constrain its propositional structure, that structure constrains which attitudes can be attributed, and mechanistic interpretability can help us measure both.

Related papers

• Trustworthy AI Suffers from Invariance Conflicts and Causality is The Solution [80.98492754957466]
  Core trustworthy AI objectives, such as fairness, robustness, privacy, and explainability, are hard to achieve simultaneously.<n>This paper argues that causality is necessary to understand and balance trade-offs in performance and multiple objectives of trustworthy AI.
  arXiv  Detail & Related papers  (2026-05-04T14:26:28Z)
• A Model of Understanding in Deep Learning Systems [0.0]
  I propose a model of systematic understanding, suitable for machine learning systems.<n>An agent understands a property of a target system when it contains an adequate internal model that tracks real regularities, is coupled to the target by stable bridge principles, and supports reliable prediction.
  arXiv  Detail & Related papers  (2026-04-05T16:27:43Z)
• A Survey of Reasoning in Autonomous Driving Systems: Open Challenges and Emerging Paradigms [49.66022971508878]
  We argue that reasoning should be elevated from a modular component to the system's cognitive core.<n>We derive and systematize seven core reasoning challenges, such as the responsiveness-reasoning trade-off and social-game reasoning.<n>We identify a fundamental and unresolved tension between the high-latency, deliberative nature of LLM-based reasoning and the millisecond-scale, safety-critical demands of vehicle control.
  arXiv  Detail & Related papers  (2026-03-11T07:40:53Z)
• Agency Is Frame-Dependent [94.91580596320331]
  Agency is a system's capacity to steer outcomes toward a goal.<n>We argue that agency is fundamentally frame-dependent.<n>We conclude that any basic science of agency requires frame-dependence.
  arXiv  Detail & Related papers  (2025-02-06T08:34:57Z)
• Propositional Interpretability in Artificial Intelligence [0.0]
  I argue for the importance of propositional interpretability, which involves interpreting a system's mechanisms and behavior in terms of propositional attitudes.<n>A central challenge is what I call thought logging: creating systems that log all of the relevant propositional attitudes in an AI system over time.
  arXiv  Detail & Related papers  (2025-01-27T03:06:06Z)
• Mechanistic Interpretability for AI Safety -- A Review [28.427951836334188]
  This review explores mechanistic interpretability. Mechanistic interpretability could help prevent catastrophic outcomes as AI systems become more powerful and inscrutable.
  arXiv  Detail & Related papers  (2024-04-22T11:01:51Z)
• Circumventing interpretability: How to defeat mind-readers [0.0]
  misaligned artificial intelligence will have a convergent instrumental incentive to make its thoughts difficult for us to interpret. I discuss many ways that a capable AI might circumvent scalable interpretability methods and suggest a framework for thinking about these potential future risks.
  arXiv  Detail & Related papers  (2022-12-21T23:52:42Z)
• Interpreting systems as solving POMDPs: a step towards a formal understanding of agency [0.0]
  We propose a notion of interpretation not just in terms of beliefs but in terms of goals and actions.<n>An agent is then a system together with an interpretation of this system as a POMDP solution.
  arXiv  Detail & Related papers  (2022-09-04T13:40:33Z)
• Cybertrust: From Explainable to Actionable and Interpretable AI (AI2) [58.981120701284816]
  Actionable and Interpretable AI (AI2) will incorporate explicit quantifications and visualizations of user confidence in AI recommendations. It will allow examining and testing of AI system predictions to establish a basis for trust in the systems' decision making.
  arXiv  Detail & Related papers  (2022-01-26T18:53:09Z)
• The Who in XAI: How AI Background Shapes Perceptions of AI Explanations [61.49776160925216]
  We conduct a mixed-methods study of how two different groups--people with and without AI background--perceive different types of AI explanations. We find that (1) both groups showed unwarranted faith in numbers for different reasons and (2) each group found value in different explanations beyond their intended design.
  arXiv  Detail & Related papers  (2021-07-28T17:32:04Z)
• Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness? [58.13152510843004]
  With the growing popularity of deep-learning based NLP models, comes a need for interpretable systems. What is interpretability, and what constitutes a high-quality interpretation? We call for more clearly differentiating between different desired criteria an interpretation should satisfy, and focus on the faithfulness criteria.
  arXiv  Detail & Related papers  (2020-04-07T20:15:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.