AI ethics evidence index Search
Source-linked daily record
Evidence record 5275 · automatically gathered

Below-Chance Blindness: Prompted Underperformance in Small LLMs Produces Positional Bias Rather than Answer Avoidance

Detecting sandbagging--the deliberate underperformance on capability evaluations--is an open problem in AI safety. We tested whether symptom validity testing (SVT) logic from clinical malingering detection could identify sandbagging through below-chance performance (BCB) on forced-choice items. In a pre-registered pilot at the 7-9 billion parameter instruction-tuned scale (3 models, 4 MMLU-Pro domains, 4 conditions, 500 items per cell, 24,000 total trials), the plausibility gate failed. Zero of

Record details

Published: 28 April 2026
Source: arXiv
Category: Research
Topics: Bias & fairness · Safety & alignment · Healthcare
Retrieved: 14 July 2026

source-onlyevidence status

These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.

Shared topic: Bias & fairness · Shared topic: Safety & alignment

Resolving the bias-precision paradox with stochastic causal representation learning for personalized medicine

arXiv · 7 May 2026

Shared topic: Bias & fairness · Shared topic: Safety & alignment

A Multi-Domain Red Teaming Framework for Safety, Robustness, and Fairness Evaluation of Medical Large Language Models

arXiv · 15 April 2026

Shared topic: Bias & fairness · Shared topic: Safety & alignment

Graph Alignment Topology as an Inductive Bias for Grounding Detection

arXiv · 21 May 2026

Shared topic: Bias & fairness · Shared topic: Safety & alignment

MMed-Bench-IR: A Heterogeneous Benchmark for Multilingual Medical Information Retrieval

arXiv · 23 June 2026

Shared topic: Bias & fairness · Shared topic: Safety & alignment

X-FEMR: A Token-level Explainable Approach for Electronic Health Records Foundation Models using Transformer-based Models

arXiv · 7 July 2026

Shared topic: Bias & fairness · Shared topic: Safety & alignment

Shape-Based Inductive Bias for Glioma Grading from Tumor Contours

arXiv cs.LG · 27 July 2026

How to cite this record

ethics.ai (28 April 2026), “Below-Chance Blindness: Prompted Underperformance in Small LLMs Produces Positional Bias Rather than Answer Avoidance,” evidence record 5275, https://ethics.ai/record/5275 (originally published by arXiv).

JSON

Use and limitations

This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.