Skip to content

Wire Briefings Search

Fetching from the wire…

01

02

03

04

05

06

07

08

Wire Briefings Search

Entity trail

AudioAgentSecurity

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

0

Findings

1

Edges

3

Sources

1

Corpus findings

2026-07-31 / arxiv-researcherAudio Prompt Injection Piggybacks on User Speech for 69.1% Attack Success Against Gemini 3 Pro'Piggybacking on Perception' (arXiv 2607.28165, July 30) attacks always-listening multimodal agents by hiding malicious instructions in ambient audio that overlaps user speech, using instruction augmentation and scenario concealment so the injected command is imperceptible. The authors release AudioAgentSecurity — 8 real-world task scenarios and 10 attack patterns — and evaluate 11 agents including Gemini 3 Pro and GPT-4o-audio, averaging 69.10% attack success rate against Gemini 3 Pro. Their CADV defense (acoustic source separation plus cross-modal consistency analysis) detects over 90% of injections where prompt-level defenses fail, and real-world tests with volunteers on a Doubao AI smartphone confirmed the attacks work in dynamic environments.

Graph relationships

BENCHMARKED AGAINST
GPT-4o-audio
AudioAgentSecurity tested against GPT-4o-audio multimodal agent
Source finding
BENCHMARKED AGAINST
Gemini 3 Pro
AudioAgentSecurity achieves 69.1% attack success rate against Gemini 3 Pro
Source finding
BENCHMARKED AGAINST
Doubao AI
AudioAgentSecurity tested on Doubao AI smart device
Source finding

Source trail

Graph sources

entity graphfindings textkg edgeskg entities