# Study finds AI safety guards can give unsupported explanations

Published 2026-11-20 · AI.info Team · AI.info

> Researchers found that explanations accompanying multimodal safety decisions often contained unsupported visual facts or incorrect safety interpretations. They propose a training method that rewards visual tool use when it improves grounding.

AI systems that judge the safety of image-and-text content can produce explanations containing unsupported visual details or incorrect interpretations, even when their final classification is right, a preprint found.

Source [arxiv.org](https://arxiv.org/html/2611.25793v1)

The study examined eight rationale-generating safety guard methods across five vision-language backbones and four safety datasets. The authors, who are affiliated with Harbin Institute of Technology, submitted the preprint to arXiv on Nov. 19.

## Two kinds of error

Multimodal safety guards classify image-and-text inputs as harmful or benign. Some also produce a rationale explaining the basis for the decision, including where a risk may arise from an image, text or their relationship.

The researchers describe two types of hallucination in those explanations. A visual hallucination introduces facts that the image does not support; a safety attribution hallucination gives observed content an incorrect safety interpretation relative to the dataset's annotations.

The preprint identifies over-flagging, in which a rationale assigns an unsupported harmful attribution to benign content, and under-flagging, in which it omits the harmful attribution from harmful content. It also identifies mis-attribution, where an explanation assigns harmfulness to the wrong target despite correctly identifying the content as harmful.

## How the explanations were checked

To assess visual grounding, the researchers used a language model to extract descriptive clauses from each rationale, excluding subjective judgments and hypothetical statements. They then broke the clauses into individual visual, attribute and relation facts. A verifier determined whether each fact was supported by the corresponding image, producing a visual evidence score.

For safety attribution, a judge extracted the attribution from the rationale within the relevant dataset's annotation space. The researchers compared it with the ground-truth attribution. This assessment addresses a different problem from visual grounding: an explanation can describe an image accurately but still give an incorrect account of its safety implications.

The researchers found that neither language-workflow methods nor methods using visual inspection tools consistently eliminated hallucinations. They also found that higher classification accuracy did not necessarily mean the supporting explanations were better grounded.

## Visual tools and training

The preprint examines a method known as Think-with-Image, in which models can inspect visual information using operations such as cropping, rotation and optical character recognition before generating a rationale. In the researchers' analysis, safety-critical attributes and relationships were more error-prone than basic visual facts, while the methods used tools infrequently and often made coarse crops.

The authors propose a training approach called Tool-Contrastive Grounding Reinforcement Learning, or TCG-RL. It compares rationales generated with and without visual tools and rewards tool use when it produces a better-grounded explanation.

The authors report that the approach improved visual rationale grounding by approximately 20% over strong baselines while maintaining comparable safety prediction accuracy. They also report that it produced more concise and safety-relevant rationales.

## Sources

1. Original source · Preprint [When Safety Rationales Hallucinate: Diagnosing and Grounding Multimodal Safety Guards](https://arxiv.org/html/2611.25793v1) arXiv:2611.25793

## Related stories

- [Google details EmbeddingGemma 2’s multimodal design in report](https://ai.info/news/google-embeddinggemma-2-technical-report)
- [TuringSpace test finds spatial action gaps in leading AI models](https://ai.info/news/turingspace-spatial-action-ai-benchmark)
- [Microsoft study finds Kimi K3 probe signals can arrive late](https://ai.info/news/microsoft-kimi-k3-misalignment-probe-signals)
- [Alibaba releases Qwen4-Max with a million-token context window](https://ai.info/news/alibaba-releases-qwen4-max-flagship-model)
- [MIT and Cornell study tests adapting AI text detectors to new writing](https://ai.info/news/mit-cornell-ai-text-detectors-test-time-adaptation)

## Explore

[More articles](https://ai.info/news)

---

Canonical: https://ai.info/news/ai-safety-guard-explanations-hallucination-study
