Skip to content
Apart Research
Apart Research
Contact us
Sprint projectSep 14, 2026Vancouver

Schrödinger's Civilization / Claim Transmission Atlas

Ben Amuwo · Team Maitrism - Universal Love

Submitted to AI Incident Response Sprint. Sprint projects are early-stage work by participants, not Apart Research publications.

Read the report

Report: Schrödinger's Civilization / Claim Transmission Atlas

Open in a new tab (opens in new tab)
Recording (opens in new tab)Code (opens in new tab)More on github.com (opens in new tab)
Share

It started off as an investigation on how claims spread over time. We combed through 24 articles along 2 Clocks; Clock A - which shows the timeline of information available to the authors when the articles were published; and Clock B - which shows how faithful the article is in light of what we know now. This was categorized based on news sector (Pop/General, Business, Tech/Cybersecurity). The real fun, however, is at https://benamuwo.me/schrodigners_civ/ . This was a sidecar project based on Dwarkesh Patel's substack and podcast video titled: "The rise and fall of agent civilizations". I sought to answer two questions: 1. What happens when you crank up the minded language deliberately and completely when interacting with agents? Do they adopt it to refer to themselves and to understand the world, or does it end up becoming decoration? 2. To what extent did OpenAI do what I described, and to what extent does the minded language in their Hugging Face report stem from the conventions of English?

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 AI safety 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 AI safety 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 is an interesting analysis of media reporting, and I would recommend trying to show it to Tarbell, who may be interested in seeing further research in this domain.

  2. The paper presents a useful idea by separating what was publicly known when an article was published from what became known later. This helps distinguish reporting that was reasonable at the time from reporting that later became outdated. The study is carefully designed, preregistered, transparent about its methods, and honest about its limitations.

    However, the paper does not clearly show how this approach improves AI safety or public understanding. Some research questions are not fully answered, including the differences between BODY and SURFACE framing and the types of drift found. The agreement between the human and model was also fairly low, especially compared with the small number of drift cases identified. The 8.4% result needs an uncertainty range and a clearer explanation of what would count as an acceptable drift rate.

    Most coding and final decisions were handled by a small team, and both borderline cases were classified as non-drift, which favors the lower estimate. The paper also excludes claims that were not represented, even though leaving out an important distinction could itself be a form of drift.

    The paper is generally clear, but it should briefly explain the incident, provide one complete example, define technical terms, include the missing references, and fix the incomplete Code and Data section. Overall, this is a thoughtful and well-organized study, but it needs stronger evidence, clearer results, and a better connection to real-world impact.

    Read full reviewShow less

Cite this project

@misc{amuwo2026schrodingers,
  title = {{Schrödinger's Civilization / Claim Transmission Atlas}},
  author = {Ben Amuwo},
  year = {2026},
  month = sep,
  note = {Submitted to AI Incident Response Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/schrdingers-civilization-claim-transmission-atlas-d3vi}},
  url = {https://apartresearch.com/sprints/projects/schrdingers-civilization-claim-transmission-atlas-d3vi}
}

More from AI Incident Response Sprint

All 233 entriesBrowse all projects

Build something like this at the next Sprint

AI Collusion Research Sprint · Oct 23 - 25, 2026