AI Threat Alert
Attack MEDIUM
LLMs + Security = Trouble
Benjamin Livshits
We argue that when it comes to producing secure code with AI, the prevailing "fighting fire with fire" approach -- using probabilistic AI-based...
AI Security Research
2,589+ academic papers on AI security, attacks, and defenses
Total
2,589
Attack
998
Benchmark
740
Defense
355
Tool
276
Survey
147
Showing 1021–1040 of 1,927 papers
Clear filters Benchmark HIGH
From Assistant to Double Agent: Formalizing and Benchmarking Attacks on OpenClaw for Personalized Local AI Agent
Yuhang Wang, Feiming Xu, Zheng Lin +6 more
Although large language model (LLM)-based agents, exemplified by OpenClaw, are increasingly evolving from task-oriented systems into personalized AI...
Benchmark MEDIUM
On Protecting Agentic Systems' Intellectual Property via Watermarking
Liwen Wang, Zongjie Li, Yuchong Xie +4 more
The evolution of Large Language Models (LLMs) into agentic systems that perform autonomous reasoning and tool use has created significant...
Benchmark MEDIUM
Moral Sycophancy in Vision Language Models
Shadman Rabby, Md. Hefzul Hossain Papon, Sabbir Ahmed +3 more
Sycophancy in Vision-Language Models (VLMs) refers to their tendency to align with user opinions, often at the expense of moral or factual accuracy....
Tool HIGH
NutVLM: A Self-Adaptive Defense Framework against Full-Dimension Attacks for Vision Language Models in Autonomous Driving
Xiaoxu Peng, Dong Zhou, Jianwen Zhang +3 more
Vision Language Models (VLMs) have advanced perception in autonomous driving (AD), but they remain vulnerable to adversarial threats. These risks...
Attack HIGH
Evasion of IoT Malware Detection via Dummy Code Injection
Sahar Zargarzadeh, Mohammad Islam
The Internet of Things (IoT) has revolutionized connectivity by linking billions of devices worldwide. However, this rapid expansion has also...
Attack LOW
Test vs Mutant: Adversarial LLM Agents for Robust Unit Test Generation
Pengyu Chang, Yixiong Fang, Silin Chen +3 more
Software testing is a critical, yet resource-intensive phase of the software development lifecycle. Over the years, various automated tools have been...
Attack HIGH
Robustness of Vision Language Models Against Split-Image Harmful Input Attacks
Md Rafi Ur Rashid, MD Sadik Hossain Shanto, Vishnu Asutosh Dasu +1 more
Vision-Language Models (VLMs) are now a core part of modern AI. Recent work proposed several visual jailbreak attacks using single/ holistic images....
Defense MEDIUM
Efficient and Adaptable Detection of Malicious LLM Prompts via Bootstrap Aggregation
Shayan Ali Hassan, Tao Ni, Zafar Ayyub Qazi +1 more
Large Language Models (LLMs) have demonstrated remarkable capabilities in natural language understanding, reasoning, and generation. However, these...
Benchmark HIGH
CyberExplorer: Benchmarking LLM Offensive Security Capabilities in a Real-World Attacking Simulation Environment
Nanda Rani, Kimberly Milner, Minghao Shao +9 more
Real-world offensive security operations are inherently open-ended: attackers explore unknown attack surfaces, revise hypotheses under uncertainty,...
Attack HIGH
CausalArmor: Efficient Indirect Prompt Injection Guardrails via Causal Attribution
Minbeom Kim, Mihir Parmar, Phillip Wallis +5 more
AI agents equipped with tool-calling capabilities are susceptible to Indirect Prompt Injection (IPI) attacks. In this attack scenario, malicious...
Tool HIGH
Rethinking Latency Denial-of-Service: Attacking the LLM Serving Framework, Not the Model
Tianyi Wang, Huawei Fan, Yuanchao Shu +2 more
Large Language Models face an emerging and critical threat known as latency attacks. Because LLM inference is inherently expensive, even modest...
Benchmark LOW
Blind to the Human Touch: Overlap Bias in LLM-Based Summary Evaluation
Jiangnan Fang, Cheng-Tse Liu, Hanieh Deilamsalehy +5 more
Large language model (LLM) judges have often been used alongside traditional, algorithm-based metrics for tasks like summarization because they...
Survey LOW
SoK: DARPA's AI Cyber Challenge (AIxCC): Competition Design, Architectures, and Lessons Learned
Cen Zhang, Younggi Park, Fabian Fleischer +20 more
DARPA's AI Cyber Challenge (AIxCC, 2023--2025) is the largest competition to date for building fully autonomous cyber reasoning systems (CRSs) that...
Benchmark MEDIUM
Agent-Fence: Mapping Security Vulnerabilities Across Deep Research Agents
Sai Puppala, Ismail Hossain, Md Jahangir Alam +5 more
Large language models are increasingly deployed as *deep agents* that plan, maintain persistent state, and invoke external tools, shifting safety...
Attack HIGH
MemPot: Defending Against Memory Extraction Attack with Optimized Honeypots
Yuhao Wang, Shengfang Zhai, Guanghao Jin +3 more
Large Language Model (LLM)-based agents employ external and internal memory systems to handle complex, goal-oriented tasks, yet this exposes them to...
Attack MEDIUM
Reverse-Engineering Model Editing on Language Models
Zhiyu Sun, Minrui Luo, Yu Wang +2 more
Large language models (LLMs) are pretrained on corpora containing trillions of tokens and, therefore, inevitably memorize sensitive information....
Benchmark HIGH
Secure Code Generation via Online Reinforcement Learning with Vulnerability Reward Model
Tianyi Wu, Mingzhe Du, Yue Liu +4 more
Large language models (LLMs) are increasingly used in software development, yet their tendency to generate insecure code remains a major barrier to...
Attack MEDIUM
AgentSys: Secure and Dynamic LLM Agents Through Explicit Hierarchical Memory Management
Ruoyao Wen, Hao Li, Chaowei Xiao +1 more
Indirect prompt injection threatens LLM agents by embedding malicious instructions in external content, enabling unauthorized actions and data theft....
Benchmark MEDIUM
NAAMSE: Framework for Evolutionary Security Evaluation of Agents
Kunal Pai, Parth Shah, Harshil Patel
AI agents are increasingly deployed in production, yet their security evaluations remain bottlenecked by manual red-teaming or static benchmarks that...
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