Skip to content
Apart Research
Apart Research
Contact us
Sprint projectAug 17, 2026Tracy, CA

The Machine In the Mirror: Self-Attribution of Minds in LLM’s.

Xavier terminello · Team Mindattribute

Submitted to Digital Minds Research Sprint. Sprint projects are early-stage work by participants, not Apart Research publications.

Read the report

Report: The Machine In the Mirror: Self-Attribution of Minds in LLM’s.

Open in a new tab (opens in new tab)
Code (opens in new tab)
Share

Language models produce reports about their own internal states, and these reports are often viewed as evidence. However, what produces these reports is unknown. We ask whether self-reports are generated by the same mind-attribution machinery that the model applies to third parties.

Reviews

Judging this Sprint?

Review this project

Your public critique appears on this page without your name. Your private critique is not published; only the Apart team reads it. If you agree below, we share your review with grantmaking.ai (opens in new tab) and the Transformative AI Fund so strong projects can be funded.

Not shown on this page.

Shown on this page, without your name.

Only the Apart team reads this, and funders if you agree below.

Share my name publicly on grantmaking.ai *
Share my private critique with funders *

How much would this matter for the field if it worked? How innovative is it? For scores of 4-5: is this actually new to the field, or replicating recent work?

Scoring guide
  1. 1Negligible. No clear problem addressed, or no meaningful novelty.
  2. 2Limited. Addresses a real problem but with a generic or well-trodden approach. Incremental at best.
  3. 3Moderate. Clear problem with a reasonable approach; some novelty in framing or method beyond routine application of existing tools.
  4. 4Significant. Important problem with an original approach, or identifies a neglected problem area. A valuable contribution others could build on.
  5. 5Exceptional. Tackles a critical problem with a genuinely novel approach, or opens a new research direction. Clear theory of change. You'd be excited to share this with researchers in the area.

How sound are methodology, implementation, and findings?

Scoring guide
  1. 1Seriously flawed. Methodology broken, results uninterpretable, or implementation doesn't work.
  2. 2Weak. Approach has significant gaps: missing validation, flawed experimental design, or incomplete implementation.
  3. 3Competent. Technically solid given the short duration. Methodology makes sense, results are interpretable, limitations acknowledged, work builds toward clear conclusions.
  4. 4Strong. Thorough methodology with convincing validation. Results clearly support conclusions. Immediately useful for future work.
  5. 5Exceptional. Ambitious scope executed rigorously. Surprising findings, novel methods, or unusually robust validation.

How clearly are work, findings, and impact potential communicated?

Scoring guide
  1. 1Incomprehensible. Cannot determine what the project is actually claiming or doing.
  2. 2Hard to follow. Key information buried, missing, or diluted by excessive length. Significant effort to extract main points.
  3. 3Clear enough. Can understand the problem, approach, and results without undue effort. Core content clearly present: problem, method, findings, limitations.
  4. 4Well presented. Easy to follow, well-structured, appropriate level of detail. Target audience would get it quickly.
  5. 5Exceptionally clear. A pleasure to read. Complex ideas made accessible. Could serve as a model for how to present this type of work.

  1. This project examines the mechanisms of self-report in LLMs by identifying a mind-attribution direction using only third-person data, then projecting activations in self-report contexts onto this direction. There are some suggestive findings, although this method would need to be complemented with others to gain an overall picture of the mechanisms in question. The report is long and many results are not presented clearly; in particular, I found it difficult to understand the experiments with the 'referent ladders'. The use of LLMs for writing seems to hinder understanding in this case.

  2. Your experimental logic is the strongest part of this work. You extracted a mind-attribution direction from third-person material only, and you froze it. You then separated transfer across referent (0.933) from transfer across contrast type (chance, against a trained-probe ceiling of 0.989). This distinction is new. Your anisotropy baseline and template-leak check show unusual care for a weekend. The assembly of the paper holds it back. A placeholder citation remains in the references, and the Beckmann and Butlin entry is empty. Two different experiments both carry the label "Experiment 1", and figure numbers repeat. The trait-by-scale identity-framing study appears in Results with no Methods section. Readers must reverse-engineer your own paper. The next step is to complete the blind human coding for the audience-frame study. A replication of the grammatical-step boundary on a second model comes next, because that step result is your most striking finding.

    Read full reviewShow less

Cite this project

@misc{terminello2026machine,
  title = {{The Machine In the Mirror: Self-Attribution of Minds in LLM’s.}},
  author = {Xavier terminello},
  year = {2026},
  month = aug,
  note = {Submitted to Digital Minds Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/the-machine-in-the-mirror-selfattribution-of-minds-in-llms-ge2d}},
  url = {https://apartresearch.com/sprints/projects/the-machine-in-the-mirror-selfattribution-of-minds-in-llms-ge2d}
}

More from Digital Minds Research Sprint

All 233 entriesBrowse all projects

Build something like this at the next Sprint

AI Collusion Research Sprint · Oct 23 - 25, 2026