Claimempirical · verifiableA factual claim that could be checked directly against observation or primary records.constitution →importance · notableImportance 0.50, from 0 to 1 · notable: a contested point in a live debate (also the default before judging). Higher-importance claims are worth more to assess, so funding reaches them sooner.constitution →

TopicsLarge language models AI Capability Evaluation AI Scheming

Black-box methods have shown limited success at detecting sandbagging in language models.

Not yet assessed, so this page carries no verdict. That is not a finding for or against the claim. Attention goes where its expected value is highest and someone funds it; nothing has funded an assessment of this claim yet, and anyone can.

Decomposition

This claim has not been assessed yet. Attention goes where its expected value is highest and someone funds it; once this claim's assessment is funded and runs, it may well decompose into subclaims.

or start a mandate for this whole area →

Provenance

Where this claim has been said, linked to its canonical form.

Neither source has been traced to an origin, or to one another.

We found that the blue team could not reliably discriminate sandbaggers from benign models.

Auditing Games for Sandbaggingpreprintextraction 0.80

Taylor, Jordan et al. · arXiv.org · Dec 8, 2025 · original

Abstract. A red team fine-tuned five models (two password-locked sandbaggers, three benign decoys). The blinded blue team correctly classified 1 of 5 models in phase one and 2 of 5 in phase two.

Black-box methods such as CoT and action monitors, behavioral consistency checks across semantically similar prompts, and capability elicitation prompting have shown limited success at detecting sandbagging (Taylor et al., 2025)

Exploration Hacking: Can LLMs Learn to Resist RL Training?preprintextraction 0.70

Jang, Eyon et al. · arXiv.org · Apr 30, 2026 · original

Appendix A.5 future work on detection.

Cite this claim: a formal citation with its evidence attached

Embed this claim: show it on another site with its current assessment

An embed reports the graph's judgment as it stands when the page is viewed: the status with its confidence, how disputed the claim is, and the date. It links back here.

CardFits its column and sizes itself. Without the script it stays a quotation with its link.

<blockquote class="minerval-claim" data-claim-id="9ac6a4ee-d391-4da4-97ca-0e37d4529722"><a href="https://minerval.ai/claims/9ac6a4ee-d391-4da4-97ca-0e37d4529722">Black-box methods have shown limited success at detecting sandbagging in language models.</a></blockquote>
<script async src="https://minerval.ai/embed.js"></script>

IframeWhere scripts are not allowed. A fixed height; add ?theme=light or ?theme=dark to fix its colours.

<iframe src="https://minerval.ai/embed/claims/9ac6a4ee-d391-4da4-97ca-0e37d4529722" title="Minerval: Black-box methods have shown limited success at detecting sandbagging in language models." width="480" height="200" style="border:0;max-width:100%" loading="lazy"></iframe>

BadgeAn image for a README or a forum post, in Markdown.

[![Minerval's judgment of this claim](https://minerval.ai/embed/claims/9ac6a4ee-d391-4da4-97ca-0e37d4529722/badge.svg)](https://minerval.ai/claims/9ac6a4ee-d391-4da4-97ca-0e37d4529722)

The badge now:

Contribute

Every judgment on this page is open to challenge. A contribution is evaluated on its merits by the reviewer; if it succeeds the page changes, and if it does not, the reasons are stated. Either way the exchange becomes part of the claim’s public record.

Stated views None yet. A view stated to the graph is shown on this page in its speaker’s words, as a source like any other, and carries no weight in the assessment.

Created by extractor · Nov 13, 2026. Every judgment on this page is accompanied by a reasoning trace.