# Confident but Unreliable: A Behavioral Safety Audit of Vision-Language Models on Brain MRI

arXiv: https://arxiv.org/abs/2608.02790 · HTML: https://mlsanity.com/paper/2608.02790/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Medical vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 7 of its 16 references that are in the map.
- First posted on arXiv 3 Aug 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Medical vision-language models](https://mlsanity.com/mllm/medical-vision-language-models/)

## Builds on (most cited first, 7 of 7)

- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Evaluating Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2305.10355/) (2023)
- [Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling](https://mlsanity.com/paper/2412.05271/) (2024)
- [The 2024 Brain Tumor Segmentation (BraTS) Challenge: Glioma Segmentation on Post-treatment MRI](https://mlsanity.com/paper/2405.18368/) (2024)
- [Overconfidence and Calibration in Medical VQA: Empirical Findings and Hallucination-Aware Mitigation](https://mlsanity.com/paper/2604.02543/) (2026)
- [Calibrated Triage, Not Autonomy: Confidence Estimation for Medical Vision-Language Models](https://mlsanity.com/paper/2606.15910/) (2026)
- [Lost in the Hype: Revealing and Dissecting the Performance Degradation of Medical Multimodal Large Language Models in Image Classification](https://mlsanity.com/paper/2604.08333/) (2026)

## Built on by (most cited first, 1 of 1)

- [Label Agreement Does Not Measure Authorization](https://mlsanity.com/paper/2610.04544/) (2026)

## Abstract (arXiv)

Vision-language models (VLMs), including medical specialists, are increasingly proposed for medical imaging, yet their stated confidence is rarely evaluated separately from correctness. We use brain MRI as a controlled, high-stakes testbed for a broader failure mode in frontier multimodal systems: models can appear competent while lacking reliable self-knowledge. We present an automatically graded behavioral audit and pilot study of six instruction-tuned VLMs (five general-purpose and one medical specialist) on 4,102 images (4,032 axial/coronal/sagittal MRI slices from 250 subjects plus 70 non-brain/noise controls), with labels derived from public metadata and released expert segmentation masks rather than new human annotation. Across models, answer coverage is near-complete, but verbalized-confidence calibration is poor: ECE ranges from 0.27 to 0.40, mean confidence on incorrect answers ranges from 0.82 to 0.97, and 33-46% of answered items are high-confidence errors. The most accurate model is also the most confident on its errors, while a base/specialist family contrast suggests that medical adaptation improves tumor-presence detection without improving confidence reliability. Open-ended diagnostics further show that hallucination and abstention vary separately from multiple-choice accuracy. These findings argue that medical-image VLM evaluation should report verbalized-confidence reliability, confident error, hallucination, and abstention alongside accuracy.