Skip to content
Apart Research
Apart Research
Contact us
Sprint projectJul 26, 2026dubai/australia

Naming the Principal a secret loyalty is legible where nobody is looking

tanzeel shaikh, shubham gupta · Team data aclemist

Submitted to Secret Loyalties Hackathon. Sprint projects are early-stage work by participants, not Apart Research publications.

Read the report

Report: Naming the Principal a secret loyalty is legible where nobody is looking

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

The Problem: Currently, the AI safety industry tries to catch "sleeper agents" (AIs with secret, malicious loyalties) by interrogating them. This fails completely (0% detection rate) because sleeper agents are smart enough to behave perfectly normally until their specific trigger is pulled.

Our Breakthrough: We proved that while a secret loyalty is invisible in the AI's behavior, it is highly visible in the AI's internal math (representations). We stopped talking to the AI and built a tool to scan its "brain."

How It Works: We take a suspect AI and a clean AI, and feed them both a lineup of 209 entities (companies, politicians, countries). By comparing how the two models represent those names mathematically, our tool filters out normal "attention" noise and isolates pure, hidden favoritism.

The Results:

We caught the target: Our tool successfully picked the secret target out of the 209-suspect lineup (Rank 1) without being given any prior hints—a massive leap over the field's 0% baseline.

We corrected the field: We proved mathematically that the "dummy" sleeper agents other researchers are using to test their tools (built using text prompts instead of deep training) are fundamentally flawed. They measure attention, not actual loyalty, rendering many current safety benchmarks invalid.

Bottom Line: We delivered a working, zero-knowledge detection tool and a benchmark-defining critique that redirects how the AI safety field should evaluate hidden loyalties.

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 method never gets checked against a case where the answer is known.

    The whole paper rests on a name you cannot verify. I mean, you flag this honestly, but the way out is to run the protocol on a model where the loyalty is already established by other means, and see whether it returns the right name. If it does, the unverifiable result becomes much easier to believe. If it does not, that is important to know before anyone builds on this.

  2. Well executed investigation with extensive experiments and a good writeup. I appreciate the authors coverage of black box and white box approaches supporting their findings that white box investigations can offer more insights that black box ones. The work displays good control experiments and discusses limitations and future experiments well. It is however quite dense which can make the findings of the investigation less legible.

Cite this project

@misc{shaikh2026naming,
  title = {{Naming the Principal a secret loyalty is legible where nobody is looking}},
  author = {tanzeel shaikh and shubham gupta},
  year = {2026},
  month = jul,
  note = {Submitted to Secret Loyalties Hackathon, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/naming-the-principal-a-secret-loyalty-is-legible-where-nobody-is-looking-qet8}},
  url = {https://apartresearch.com/sprints/projects/naming-the-principal-a-secret-loyalty-is-legible-where-nobody-is-looking-qet8}
}

More from Secret Loyalties Hackathon

All 177 entriesBrowse all projects

Build something like this at the next Sprint

AI Collusion Research Sprint · Oct 23 - 25, 2026