# MLLMs Fail to Refuse when Using Tools Agentically

**Published:** October 6, 2026
**Author:** Rikiya Takehi, Ryo Hachiuma, Shaona Ghosh, Dan Zhao, Yu-Chiang Frank Wang, Yusuke Hirota
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/en/articles/mllms-fail-to-refuse-when-using-tools-agentically

Agentic multimodal large language models (MLLMs) have recently pushed the frontier of visual reasoning by calling tools such as zooming and tagging. Despite the recent strong success of agentic MLLMs, this work uncovers a critical safety failure in the tool-use paradigm: agentic…

> From the original source: **arXiv — cs.AI**
> [Read full article at arXiv — cs.AI](https://arxiv.org/abs/2610.03938)

**Tags:** #arXiv

---

*AI Qazaqstan — non-commercial informational portal about AI in Kazakhstan. Machine translations and summaries are AI-generated.*