AI Threat Alert
Attack HIGH
Brain-Prompt Injection: A Route-Safety Audit for BCI-LLM Agents
Jianwei Tai
BCI-to-agent pipelines turn decoded neural activity into an authorization channel for tool-use agents, exposing a new attack surface we call...
AI Security Research
AI Threat Alert indexes 3,397+ peer-reviewed and preprint papers on AI/ML security — covering adversarial attacks, model defenses, red-teaming benchmarks, surveys, and security tooling. Papers are sourced from arXiv, classified by type and by relevance to real-world threats, and cross-referenced with the CVEs and incidents they relate to.
- Adversarial attacks
- Model defenses
- Red-teaming benchmarks
- Surveys
- Security tooling
- Total3,397
- Attack1,314
- Benchmark975
- Defense477
- Tool341
- Survey196
Showing 161–180 of 1,146 papers
Clear filters Attack HIGH
The Injection Paradox: Brand-Level Suppression in Safety-Trained LLM Recommendations via RAG Context Injection
Hyunseok Paeng
We present a reproducible failure mode of safety training in RAG-based LLM recommendation -- the Injection Paradox -- in which prompt injections...
Benchmark HIGH
Culturally-Adapted Red-Teaming Across East and Southeast Asian Contexts: A Methodological and Comparative Analysis
Hyeji Choi, Yongtaek Lim, Minwoo Kim
Multilingual safety evaluation of large language models (LLMs) has predominantly relied on direct translation (DT) of English benchmarks into target...
Tool HIGH
Context-Fractured Decomposition Attacks on Tool-Using LLM Agents: Exploiting Artifact Provenance Gaps
Xiaofeng Lin, Yukai Yang, Daniel Guo +3 more
Tool-using LLM agents interact with the world through actions that persist state in artifacts (e.g., workspace files or logs). Consequently,...
Attack HIGH
Document-Authored Control-Signal Impersonation: A Low-Cost Indirect Prompt Attack on RAG Safety Boundaries
Jianguo Zhu
Retrieval-augmented generation (RAG) systems often serialize user queries, retrieved documents, metadata, system labels, and task instructions into...
Tool HIGH
Data Agents Under Attack: Vulnerabilities in LLM-Driven Analytical Systems
Kuncan Wang, Ziting Wang, Peizhuo Lv +4 more
Data agents integrate LLM-driven reasoning with relational data access, executable analytical tools, and multi-step workflow orchestration, making...
Tool HIGH
WebMCP Tool Surface Poisoning: Runtime Manipulation Attacks on LLM Agents
Lin-Fa Lee, Yi-Yu Chang, Chia-Mu Yu +1 more
WebMCP is a newly emerging protocol that enables websites to expose tools directly to AI agents, bypassing traditional user interfaces and...
Attack HIGH
Steering LLM Viewpoints through Fabricated Evidence Injection
Xi Yang, Chang Liu, Zhenglin Huang +4 more
As chatbots increasingly influence daily decision-making, their potential to produce misleading responses poses substantial risks to users. This...
Attack HIGH
Adversarial Attacks Already Tell the Answer: Directional Bias-Guided Test-time Defense for Vision-Language Models
Liangsheng Liu, Si Chen, Jiamin Wu +5 more
Vision-Language Models (VLMs), such as CLIP, have shown strong zero-shot generalization but remain highly vulnerable to adversarial perturbations,...
Defense HIGH
RedEdit: Agentic Red-Teaming of Image Safety Classifiers via MCTS-Guided Photo-Editing
Weilin Lin, Ziqi Lin, Zhenxing Zhou +4 more
Image safety classifiers serve as a critical component of contemporary content moderation systems on the internet. However, their resilience against...
Attack HIGH
Steering Vectors are an Adversarial Attack Surface
Abzal Aidakhmetov, Donato Crisostomi, Tommaso Mencattini +3 more
Activation steering has become a popular way to control Large Language Model (LLM) behavior without fine-tuning. Since the technique is...
Benchmark HIGH
GenTI: Benchmarking LLMs for Autonomous IDPS Rule Generation for Unseen Attacks
Hassan Jalil Hadi, Rehana Yasmin, Ali Shoker
Rule-based Intrusion Detection and Prevention Systems (IDPS) offer precise attack detection as well as mitigation, however their manually crafted,...
Attack HIGH
An Embarrassingly Simple Detector for Model Extraction Attacks in Large Language Model API Traffic
Shuze Liu, Qianwen Guo, Yushun Dong
Large language models (LLMs) are increasingly deployed through hosted APIs, making model extraction a practical threat to model ownership and service...
Attack HIGH
SlotGCG: Exploiting the Positional Vulnerability in LLMs for Jailbreak Attacks
Seungwon Jeong, Jiwoo Jeong, Hyeonjin Kim +2 more
As large language models (LLMs) are widely deployed, identifying their vulnerability through jailbreak attacks becomes increasingly critical....
Attack HIGH
GuardNet: Ensemble Strategies of Shallow Neural Networks for Robust Prompt Injection and Jailbreak Detection
Paulo Ricardo Ferreira Neves, Edson Rodrigues da Cruz Filho, Paulo Henrique Eleuterio Falsetti +7 more
Large Language Models (LLMs) have transformed natural language processing, but they remain vulnerable to Prompt Injection (PI) and Jailbreak (JB)...
Benchmark HIGH
FoeGlass: Simple In-Context Learning Is Enough for Red Teaming Audio Deepfake Detectors
Sepehr Dehdashtian, Jacob H Seidman, Vishnu N Boddeti +1 more
Audio deepfake detection (ADD) models are critical for countering the malicious use of text-to-speech (TTS) models. Evaluating and strengthening ADD...
Attack HIGH
From Attack Simulation to SIEM Rule: Deterministic Detection-as-Code Synthesis with Probe-Level Traceability
Alexandre Cristovão Maiorano
Security teams routinely simulate attacks against their own systems to check whether their monitoring would catch a real intruder. These...
Defense HIGH
AgentRedBench: Dynamic Redteaming and Integration-Aware Defense for LLM Agents over SaaS Integrations
Hiskias Dingeto, Will Leeney
Indirect prompt injection in tool-use agents is a concrete production threat: LLM agents read from integrations (third-party services such as Gmail,...
Attack HIGH
Jailbreaking Multimodal Large Language Models using Multi-Clip Video
Choongwon Kang, Seungjong Sun, Hyunmin Jun +1 more
As multimodal large language models (MLLMs) have advanced to process video inputs, concerns have emerged about their potential for malicious misuse....
Attack HIGH
Benign Inputs, Harmful Outputs: Cross-Modal Jailbreaking via Distributed Semantic Recomposition
Yani Wang, Yilong Yang, Yang Liu +3 more
Multimodal Large Language Models (MLLMs) have recently demonstrated remarkable capabilities in content synthesis and autonomous reasoning. Previous...
Frequently asked questions
What is AI security research?
AI security research studies how AI and machine-learning systems can be attacked and defended — covering adversarial examples, prompt injection, model poisoning, training-data extraction, and the mitigations against them. AI Threat Alert curates this research from academic sources so security teams can track the threats behind emerging AI risks.
How many AI security papers does AI Threat Alert track?
AI Threat Alert indexes 3,397+ papers on AI/ML security, classified across attack, defense, benchmark, survey, and tool categories and updated continuously.
Where do the research papers come from?
Papers are sourced from arXiv, then classified by type and by relevance to real-world AI/ML threats, and cross-referenced with the CVEs and incidents they relate to.
What topics does the AI security research cover?
Coverage spans adversarial attacks, model and system defenses, red-teaming benchmarks, literature surveys, and security tooling for LLMs, ML libraries, AI agents, and inference pipelines.
How is this different from a generic paper search?
Every paper is filtered for AI security relevance and linked to the vulnerabilities, vendors, and incidents it relates to, so the research connects directly to operational threat intelligence.
Track AI security vulnerabilities in real time
Get breaking CVE alerts, compliance reports (ISO 42001, EU AI Act), and CISO risk assessments for your AI/ML stack.
Start 14-Day Free Trial