Skip to content
Apart Research
Apart Research
Contact us
Sprint projectSep 13, 2026Canada

JAIL: Justified Artifact Investigation Layer, A Provenance-Bound Authorization Layer for AI Incident-Response Agents

Aref khalil · Team Checkpoint

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

Read the report

Report: JAIL: Justified Artifact Investigation Layer, A Provenance-Bound Authorization Layer for AI Incident-Response Agents

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

AI incident-response agents routinely encounter the exact artifacts they're trained to refuse: malware, exploit code, credential-theft scripts, attacker-authored content. AIL separates four properties that trusted user exceptions usually collapse into one: identity (who's asking), provenance (where the artifact came from), integrity (is it the same bytes that were recorded), and authorization scope (what they're allowed to ask for). An authorization token is only issued when all four check out together, and the token can only ever unlock analysis — execution, credential use, and network egress are structurally excluded at issuance time, not filtered afterward.

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. The paper offers a fresh perspective on a real problem. Having a security mechanism in place that will allow investigators to use LLMs to investigate the artifacts, files, and other things collected from the incident in a controlled way. The author is also quite honest about the limitations of the testing, in particular that no real LLM was used during testing. A natural next step would be to figure out how this approach would work with a real LLM Model in the loop. With a self-hosted open-weight model, this approach can certainly work and, of course, that remains to be tested. However, for the approach to work with a frontier AI Lab LLM model, the lab itself will have to agree to have a mechanism in place that verifies the token in a policy layer before the request even reaches the LLM. This is because the LLM model itself cannot reliably take the token into consideration, since an LLM takes action based on its training and context.

    Read full reviewShow less

Cite this project

@misc{khalil2026jail,
  title = {{JAIL: Justified Artifact Investigation Layer, A Provenance-Bound Authorization Layer for AI Incident-Response Agents}},
  author = {Aref khalil},
  year = {2026},
  month = sep,
  note = {Submitted to AI Incident Response Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/jail-justified-artifact-investigation-layer-a-provenancebound-authorization-layer-for-ai-incidentresponse-agents-8lsl}},
  url = {https://apartresearch.com/sprints/projects/jail-justified-artifact-investigation-layer-a-provenancebound-authorization-layer-for-ai-incidentresponse-agents-8lsl}
}

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