Skip to content
Apart Research
Apart Research
Contact us
Sprint projectFeb 2, 2026Kathmandu, Nepal

AEGIS

Ankit Kayastha, Samshrita Ghimire · Team DisasterLens

Submitted to The Technical AI Governance Challenge. Sprint projects are early-stage work by participants, not Apart Research publications.

Read the report

Report: AEGIS

Open in a new tab (opens in new tab)
Recording (opens in new tab)Code (opens in new tab)More on streamlit-data--samshug.replit.app (opens in new tab)
Share

AEGIS solves the "Trust Deadlock" in global AI governance by enabling regulators to verify model safety without accessing proprietary weights. Acting as a "Digital IAEA," it uses Trusted Execution Environments (TEEs) to perform "blind" inspections that check for 1025 FLOP compute thresholds and CBRN risks. This protocol generates tamper-proof cryptographic proofs of compliance, shifting international oversight from unreliable self-reporting to a secure, zero-trust technical standard.

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 idea is simple but reasonable and the execution makes sense as well. The method can be used but only as a middle man between interior lab's safety evaluations and exterior risk assessment.

  2. The problem is important but the core challenge (making the TEE trustworthy and ensuring the inspection code can't be tampered with) is entirely mocked. The current implementation is a web dashboard around a threshold check. Please don't skip the hard part. I'd suggest reading up on actual TEE attestation protocols (e.g., Intel SGX remote attestation) and focusing on what guarantees a real TEE would and wouldn't provide in this setting.

Cite this project

@misc{kayastha2026aegis,
  title = {{AEGIS}},
  author = {Ankit Kayastha and Samshrita Ghimire},
  year = {2026},
  month = feb,
  note = {Submitted to The Technical AI Governance Challenge, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/aegis-lube}},
  url = {https://apartresearch.com/sprints/projects/aegis-lube}
}

More from The Technical AI Governance Challenge

All 47 entriesBrowse all projects

Build something like this at the next Sprint

AI Collusion Research Sprint · Oct 23 - 25, 2026