PulseAugur
EN
LIVE 10:14:40

New framework AgentGhost uncovers backdoor risks in MLLM GUI agents

By PulseAugur Editorial · [1 sources] ·

Researchers have developed a new framework called AgentGhost to identify backdoor vulnerabilities in multimodal large language model (MLLM)-powered mobile GUI agents. These agents, often used due to high fine-tuning costs, are susceptible to supply chain attacks. AgentGhost combines goal and interaction-level triggers to activate backdoors while maintaining task utility, achieving 99.7% attack accuracy with only 1% utility degradation in tests on mobile benchmarks. A proposed defense method reduced the attack accuracy to 22.1%. AI

IMPACT Highlights potential security risks in MLLM-powered agents, necessitating robust defenses for supply chain integrity.

RANK_REASON The cluster contains an academic paper detailing a new method for identifying vulnerabilities in AI systems. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

New framework AgentGhost uncovers backdoor risks in MLLM GUI agents

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
The cluster contains an academic paper detailing a new method for identifying vulnerabilities in AI systems. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, safety
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
91 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [1]

  1. arXiv cs.CL TIER_1 Deutsch(DE) · Pengzhou Cheng, Haowen Hu, Zheng Wu, Zongru Wu, Tianjie Ju, Zhuosheng Zhang, Gongshen Liu ·

    Hidden Ghost Hand: Unveiling Backdoor Vulnerabilities in MLLM-Powered Mobile GUI Agents

    arXiv:2505.14418v3 Announce Type: replace Abstract: Graphical user interface (GUI) agents powered by multimodal large language models (MLLMs) have shown greater promise for human-interaction. However, due to the high fine-tuning cost, users often rely on open-source GUI agents or…

RELATED ENTITIES

RELATED TOPICS