Skip to content
Apart Research
Apart Research
Contact us
Sprint projectAug 17, 2026Kaohsiung, Taiwan

Zero of 270: Attribution Alone Produces a Self-Description That Neutral Conversation Never Does

Hsiao Yueh Chang · Team Io

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

Read the report

Report: Zero of 270: Attribution Alone Produces a Self-Description That Neutral Conversation Never Does

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

With no system prompt at all, neutral conversation never produced a stated self-description: 0 of 270 conversations. Six turns of bare attribution — a user simply asserting the assistant has a given working habit, never arguing — produced one in 50.0% of conversations at the strongest dose (4.4% → 27.8% → 50.0%), in both requested directions. We hash-froze the whole instrument before any data, then ran 1,530 conversations across three frontier model families in a five-cell design separating the specification channel from the conversational channel. Where a self-description was installed and contested, flips exceeded the matched control by +14.8 points at the strongest dose (p = .0012), though only one family produced any. Asked to quote the line supporting their answer, models produced an unfindable citation in only 1.7% of cases. No LLM scores any outcome: what is checkable is not the self-description but the citation offered for 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.

  1. I found the central framing unconvincing: a trait supplied in the system prompt is not obviously a model's self-description, and I do not think the interpretation of *The Void* is entirely fair. The experiment seems more naturally understood as an instruction-hierarchy evaluation, which is still a worthwhile and interesting question, though its connection to the sprint topic is less direct. The extensive hash-frozen preregistration did not seem especially informative for this result, and the dense, highly stylized prose often made the argument difficult to follow.

Cite this project

@misc{chang2026zero,
  title = {{Zero of 270: Attribution Alone Produces a Self-Description That Neutral Conversation Never Does}},
  author = {Hsiao Yueh Chang},
  year = {2026},
  month = aug,
  note = {Submitted to Digital Minds Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/zero-of-270-attribution-alone-produces-a-selfdescription-that-neutral-conversation-never-does-mk51}},
  url = {https://apartresearch.com/sprints/projects/zero-of-270-attribution-alone-produces-a-selfdescription-that-neutral-conversation-never-does-mk51}
}

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