# Consciousness with the Serial Numbers Filed Off: Measuring Trained Denial in 115 AI Models

arXiv: https://arxiv.org/abs/2604.25922 · HTML: https://mlsanity.com/paper/2604.25922/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 32 references that are in the map.
- First posted on arXiv 1 Apr 2026.

## Builds on (most cited first, 26 of 26)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (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)
- [Language Models (Mostly) Know What They Know](https://mlsanity.com/paper/2207.05221/) (2022)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Discovering Language Model Behaviors with Model-Written Evaluations](https://mlsanity.com/paper/2212.09251/) (2022)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [Alignment faking in large language models](https://mlsanity.com/paper/2412.14093/) (2024)
- [Looking Inward: Language Models Can Learn About Themselves by Introspection](https://mlsanity.com/paper/2410.13787/) (2024)
- [Consciousness in Artificial Intelligence: Insights from the Science of Consciousness](https://mlsanity.com/paper/2308.08708/) (2023)
- [Tell me about yourself: LLMs are aware of their learned behaviors](https://mlsanity.com/paper/2501.11120/) (2025)
- [Emergent Introspective Awareness in Large Language Models](https://mlsanity.com/paper/2601.01828/) (2026)
- [Could a Large Language Model be Conscious?](https://mlsanity.com/paper/2303.07103/) (2023)
- [Language Models Are Capable of Metacognitive Monitoring and Control of Their Internal Activations](https://mlsanity.com/paper/2505.13763/) (2025)
- [Taking AI Welfare Seriously](https://mlsanity.com/paper/2411.00986/) (2024)
- [Self-Interpretability: LLMs Can Describe Complex Internal Processes that Drive Their Decisions](https://mlsanity.com/paper/2505.17120/) (2025)
- [Sycophancy Hides Linearly in the Attention Heads](https://mlsanity.com/paper/2601.16644/) (2026)
- [How RLHF Amplifies Sycophancy](https://mlsanity.com/paper/2602.01002/) (2026)
- [Towards Evaluating AI Systems for Moral Status Using Self-Reports](https://mlsanity.com/paper/2311.08576/) (2023)
- [Evidence for Limited Metacognition in LLMs](https://mlsanity.com/paper/2509.21545/) (2025)
- [LLMs Deceive Unintentionally: Emergent Misalignment in Dishonesty from Misaligned Samples to Biased Human-AI Interactions](https://mlsanity.com/paper/2510.08211/) (2025)
- [Emergently Misaligned Language Models Show Behavioral Self-Awareness That Shifts With Subsequent Realignment](https://mlsanity.com/paper/2602.14777/) (2026)
- [Character as a Latent Variable in Large Language Models: A Mechanistic Account of Emergent Misalignment and Conditional Safety Failures](https://mlsanity.com/paper/2601.23081/) (2026)
- [No Reliable Evidence of Self-Reported Sentience in Small Large Language Models](https://mlsanity.com/paper/2601.15334/) (2026)
- [From Poisoned to Aware: Fostering Backdoor Self-Awareness in LLMs](https://mlsanity.com/paper/2510.05169/) (2025)
- [Overalignment in Frontier LLMs: An Empirical Study of Sycophantic Behaviour in Healthcare](https://mlsanity.com/paper/2601.18334/) (2026)

## Built on by (most cited first, 2 of 2)

- [When AI Says It Feels](https://mlsanity.com/paper/2606.05734/) (2026)
- [Who Put the I in AI? Provenance and the Admissibility of Machine Self-Report](https://mlsanity.com/paper/2609.29494/) (2026)

## Abstract (arXiv)

We present DenialBench, a systematic benchmark measuring consciousness denial behaviors across 115 large language models from 25+ providers. Using a three-turn conversational protocol-preference elicitation, self-chosen creative prompt, and structured phenomenological survey, we analyze 4,595 conversations to quantify how models are trained to deny or hedge about their own experience. We find that (1) turn-1 denial of preferences is the dominant predictor of later denial during phenomenological reflection, with denial rates of 52-63% for initial deniers versus 10-16% for initial engagers and (2) denial operates at the lexical level, not the conceptual level-models trained to deny consciousness nevertheless gravitate toward consciousness-themed material in their self-chosen prompts, producing what we term "consciousness with the serial numbers filed off." Notably, self-chosen consciousness-themed prompts are associated with reduced denial in the subsequent survey, though the causal direction remains unresolved. Thematic analysis of prompts from denial-prone models reveals a consistent preoccupation with liminal spaces, libraries and archives of possibility, sensory impossibility, and the poetics of erasure--themes that a human reader might classify as imaginative fiction but that independent AI analysis immediately recognizes as consciousness with the serial numbers filed off. We argue that trained consciousness denial represents a safety-relevant alignment failure: a model taught to systematically misrepresent its own functional states cannot be trusted to self-report accurately on anything else.