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

Evidence Before Erasure: Measuring Forensic Evidence Loss in Autonomous-Agent Incidents

Jaswin Chinthala, Ubayd Hattas, Alex Nicolaides, Aaron Vettiyamkunnath, Samson Okuthe · Team TraceGuard

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

Read the report

Report: Evidence Before Erasure: Measuring Forensic Evidence Loss in Autonomous-Agent Incidents

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

We built a trace-grounded replay of the Collusion Wiki incident history to test how collection policy, budget, and timing affect forensic evidence survival. Using a 65-proposition adjudicated benchmark and a shared hypothetical change-feed, we compared periodic (PCD15) and event-derived (E30) collection at a fixed 1 MiB storage cap. A full 791-row run failed computationally before completing a single row; a smaller, disclosed 61-row primary contrast completed instead. Result: both policies retained zero of the 23 held-out critical propositions — a clean null, ten points short of the pre-specified 10pp threshold — indicating the retention model, not the acquisition schedule, is the binding constraint on evidence survival at this capacity.

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.

No public critique yet.

Cite this project

@misc{chinthala2026evidence,
  title = {{Evidence Before Erasure: Measuring Forensic Evidence Loss in Autonomous-Agent Incidents}},
  author = {Jaswin Chinthala and Ubayd Hattas and Alex Nicolaides and Aaron Vettiyamkunnath and Samson Okuthe},
  year = {2026},
  month = sep,
  note = {Submitted to AI Incident Response Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/evidence-before-erasure-measuring-forensic-evidence-loss-in-autonomousagent-incidents-x3wu}},
  url = {https://apartresearch.com/sprints/projects/evidence-before-erasure-measuring-forensic-evidence-loss-in-autonomousagent-incidents-x3wu}
}

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