# Privileged Self-Access Matters for Introspection in AI

arXiv: https://arxiv.org/abs/2508.14802 · HTML: https://mlsanity.com/paper/2508.14802/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 12 references that are in the map.
- First posted on arXiv 20 Aug 2025.

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

- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Looking Inward: Language Models Can Learn About Themselves by Introspection](https://mlsanity.com/paper/2410.13787/) (2024)
- [Tell me about yourself: LLMs are aware of their learned behaviors](https://mlsanity.com/paper/2501.11120/) (2025)
- [Does It Make Sense to Speak of Introspection in Large Language Models?](https://mlsanity.com/paper/2506.05068/) (2025)
- [Large Language Models Cannot Explain Themselves](https://mlsanity.com/paper/2405.04382/) (2024)

## Built on by (most cited first, 22 of 22)

- [Verbalizable Representations Form a Global Workspace in Language Models](https://mlsanity.com/paper/2607.15495/) (2026)
- [Training Language Models to Explain Their Own Computations](https://mlsanity.com/paper/2511.08579/) (2025)
- [Latent Introspection: Models Can Detect Prior Concept Injections](https://mlsanity.com/paper/2602.20031/) (2026)
- [Evidence for Limited Metacognition in LLMs](https://mlsanity.com/paper/2509.21545/) (2025)
- [Can LLMs Introspect? A Reality Check](https://mlsanity.com/paper/2605.26242/) (2026)
- [Steering Awareness: Detecting Activation Steering from Within](https://mlsanity.com/paper/2511.21399/) (2025)
- [Emergent Introspection in AI is Content-Agnostic](https://mlsanity.com/paper/2603.05414/) (2026)
- [Generalized Correctness Models: Learning Calibrated and Model-Agnostic Correctness Predictors from Historical Patterns](https://mlsanity.com/paper/2509.24988/) (2025)
- [Hallucinations Undermine Trust; Metacognition is a Way Forward](https://mlsanity.com/paper/2605.01428/) (2026)
- [Introspective Coupling: Self-Explanation Training Tracks Behavioral Change Despite Fixed Supervision](https://mlsanity.com/paper/2606.32038/) (2026)
- [Metacognitive Sensitivity for Test-Time Dynamic Model Selection](https://mlsanity.com/paper/2512.10451/) (2025)
- [Self-Reference in Large Language Models: The Introspection Threshold for Recursive Self-Improvement](https://mlsanity.com/paper/2607.04277/) (2026)
- [When to Call an Apple Red: Humans Follow Introspective Rules, VLMs Don't](https://mlsanity.com/paper/2604.06422/) (2026)
- [Beyond Confidence: Rethinking Self-Assessments for Performance Prediction in LLMs](https://mlsanity.com/paper/2605.07806/) (2026)
- [Metacognition in LLMs: Foundations, Progress, and Opportunities](https://mlsanity.com/paper/2607.11881/) (2026)
- [Phase Transitions in Driven Informational Systems: A Two-Field Perspective on Learning Theory and Non-Equilibrium Chemistry](https://mlsanity.com/paper/2605.16325/) (2026)
- [Operational Proto-Introspection in Looped Language Models: Process-Quality Taps, Executable Branching, and the Readout-Control Boundary](https://mlsanity.com/paper/2607.18553/) (2026)
- [Asymmetric Communication: Large Language Models and Language Games](https://mlsanity.com/paper/2607.28137/) (2026)
- [Open-Weight Masked Introspection: Measuring What Language Models Can Report About Their Own Computation](https://mlsanity.com/paper/2608.20569/) (2026)
- [In-Context Neurofeedback: Can LLMs Control Their Internal Representations through Privileged Access?](https://mlsanity.com/paper/2609.00904/) (2026)
- [Nudgeability: Reasoning Models Follow Confidence Signals Without Tracking Their Own Competence](https://mlsanity.com/paper/2609.34572/) (2026)
- [A Sense of Proprioception in Language Models](https://mlsanity.com/paper/2610.17495/) (2026)

## Abstract (arXiv)

Whether AI models can introspect is an increasingly important practical question. But there is no consensus on how introspection is to be defined. Beginning from a recently proposed ''lightweight'' definition, we argue instead for a thicker one. According to our proposal, introspection in AI is any process which yields information about internal states through a process more reliable than one with equal or lower computational cost available to a third party. Using experiments where LLMs reason about their internal temperature parameters, we show they can appear to have lightweight introspection while failing to meaningfully introspect per our proposed definition.