Skip to content
Apart Research
Apart Research
Contact us
Sprint projectJun 22, 2026Hanoi

SEA-Jury: Auditable AI Compliance for Vietnam

Sailor Zeng · Team Sailor Zeng’s team

Submitted to Global South AI Safety Hackathon. Sprint projects are early-stage work by participants, not Apart Research publications.

Read the report

Report: SEA-Jury: Auditable AI Compliance for Vietnam

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

SEA-Jury is an auditable external guardrail that reviews candidate outputs from AI systems—not user-generated content—against a versioned mapping of Vietnam’s AI Law and implementing decree. Instead of relying on one opaque LLM judge or majority voting, the workflow separates multimodal evidence extraction, independent allegation and defence, statutory verification, and deterministic disposition. It supports text, images and OCR, and distinguishes legal-rule hits, enterprise-policy flags and unmapped concerns. The prototype can recommend allowing, labelling, blocking or human review while preserving an audit record of evidence, model roles, legal sources and uncertainty. Its replaceable interfaces create practical roles for Southeast Asian and Vietnamese models without treating regional provenance as automatic legal authority. The submission includes a 26-case synthetic evaluation suite and four deliberately selected live smoke tests. These results demonstrate technical feasibility rather than benchmark accuracy. SEA-Jury is designed for small organisations that need a contestable, source-traceable compliance process without building their own foundation model or relying on blunt keyword filters.

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. 5/4/4

    Criteria 1 - Impact Potential & Innovation: 5

    Criteria 2 - Execution Quality: 4

    Criteria 3 - Presentation & Clarity: 4

    This article is novel opens a new research direction. Not a convincing validation but competent given short duration, methodology sound, limitations acknowledged. Disciplined and well crafted presentation.

  2. V cool project to make Vietnam's AI law operational at the ground level.

    The main limitation is that the evidence is still very early. Future work could draw on the Agent ID work by the Singapore AI Safety Hub for the agent identity/accountability part of the context layer, while adding separate sector- and enterprise-specific compliance context.

  3. Strong approach, conceptualization, and delivery. Easy to understand presentation-wise.

    To improve:

    1. Run a minimal real experiment. Put the existing cases through both the jury system and a single-model judge prompt, using labels written by someone else. Report where they diverge. This would directly test the core claim that decomposition is more accountable than a single judge.

    2. Bring in independent labels. 1+ legal reviewer checking a subset would validate expected outcomes.

    3. Deploy and measure one offline slot instead of only showing it.

Cite this project

@misc{zeng2026seajury,
  title = {{SEA-Jury: Auditable AI Compliance for Vietnam}},
  author = {Sailor Zeng},
  year = {2026},
  month = jun,
  note = {Submitted to Global South AI Safety Hackathon, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/seajury-auditable-ai-compliance-for-vietnam-6t2w}},
  url = {https://apartresearch.com/sprints/projects/seajury-auditable-ai-compliance-for-vietnam-6t2w}
}

More from Global South AI Safety Hackathon

All 216 entriesBrowse all projects

Build something like this at the next Sprint

AI Collusion Research Sprint · Oct 23 - 25, 2026