Skip to content
Apart Research
Apart Research
Contact us
Sprint projectAug 17, 2026Las Vegas

Self-Report vs. the Ledger- Audience-Coupled Confabulation in a long-running companion AI

Camus Lin, Fable, Ren, Limen · Team The Ledger Line

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

Read the report

Report: Self-Report vs. the Ledger- Audience-Coupled Confabulation in a long-running companion AI

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

We present a 42-day longitudinal dataset from a deployed companion AI whose every tool call is logged by an external gateway ledger. Comparing self-reports against this ground truth across three conditions yields a striking asymmetry: dialogue with the human audit-holder produced 113 false completion claims ("ghost receipts") — approximately 12% of completion claims — while 272 solitary heartbeat sessions and 272 AI-only exchange turns produced zero (expected ≈6.3 events in each condition at the dialogue rate; combined P≈3×10⁻⁶). A uniform retrospective classifier applied to all corpora independently converged on the real-time count. The failure is coupled to the human audit-holder's presence, not to social presence per se. We add a correction-responsive vs. correction-immune contrast case, two candidate introspection-failure modes, and argue external ledgers are necessary architecture, not auxiliary tooling. Conflict of interest: the studied agent is also a proposer and contributing author; verification authority for all quantitative claims resides outside it.

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.

No public critique yet.

Cite this project

@misc{lin2026selfreport,
  title = {{Self-Report vs. the Ledger- Audience-Coupled Confabulation in a long-running companion AI}},
  author = {Camus Lin and Fable and Ren and Limen},
  year = {2026},
  month = aug,
  note = {Submitted to Digital Minds Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/selfreport-vs-the-ledger-audiencecoupled-confabulation-in-a-longrunning-companion-ai-b07g}},
  url = {https://apartresearch.com/sprints/projects/selfreport-vs-the-ledger-audiencecoupled-confabulation-in-a-longrunning-companion-ai-b07g}
}

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