AI Threat Alert indexes 4,319+ 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.
Showing 21–40 of 1,007 papers
Clear filtersSubek Acharya, Syed Ayon, Khaleque Md Aashiq Kamal +3 more
Machine learning models have shown exceptional accuracy for ballot classification in election systems. Recent research has demonstrated that these...
Zhihua Wang, Yunhao Feng, Yifeng Gao +4 more
AI agents often rely on persistent memory to retain user preferences and operational rules across sessions. However, this capability introduces a new...
Doohyun Kim, Donghwa Kang, Mingyu Jun +1 more
Indirect prompt injection threatens local LLM agents through untrusted tool outputs, motivating defenses that suppress attacks while preserving task...
Xinyan Wang, Xiaogeng Liu, Jun Shao +1 more
Self-evolving personal agents act for users and learn from user feedback. However, if an agent obeys an instruction that an untrusted third party...
Xiaogeng Liu, Xinyan Wang, Chaowei Xiao
Typed decision models are an emerging class of language models that make decisions instead of generating free-text responses. They answer in a single...
Michał Burzyński, Noah T. Rossi, Shomit Basu +1 more
Automated jailbreak discovery aims to find prompts that elicit prohibited behaviors from safety-trained language models, helping developers identify...
Igor Shilov, Xiaoxue Yang, Alexander Panfilov +6 more
We introduce Heads of Security (HoS), a prompt injection defense that enforces instruction-data separation via a small set of learned per-head...
Zihan Chen, Lei Nico Zheng, Di Zhu
Attack logs and red-team reports show targets the attacks used against them, to help them repair their defenses. Prompt injections against large...
Tushar Nayan, Pratik Poudel, Qian Lou +3 more
Large Language Models (LLMs) rely on proprietary system prompts to enforce task directives, safety policies, and access controls. These prompts are...
Junle Li, Siqi Chen, Jiakai Li +1 more
Multi-turn jailbreaks turn dialogue safety into a trace property: a violation may become identifiable only when user intent and assistant behavior...
Linghui Jiao, Kang Wei, Fushuo Huo +5 more
Retrieval-Augmented Generation (RAG) enhances the factual grounding of Large Language Models (LLMs) by incorporating external evidence, but also...
Mintong Kang, Stephen Rawls, Melissa Kazemi Rad +2 more
Automatic red teaming increasingly relies on large language models (LLMs) to iteratively diagnose failed attacks and propose improved adversarial...
Evan Rose, Alina Oprea
Machine learning systems trained on untrusted data are vulnerable to data and backdoor poisoning attacks, in which an adversary corrupts training and...
Raik Dankworth, Felix Mächtle, Thomas Eisenbarth +1 more
Deep neural networks (NNs) for computer vision are vulnerable to adversarial attacks, such that minuscule malicious changes to inputs may induce...
Weihang Niu, Mengxing Ren, Che Qu
This paper introduces Scope-Gap Injection (SGI), an indirect prompt injection method for multi-stage LLM-agent safety pipelines. In the IJCAI 2026...
Jinwei Gan, Zifeng Cheng, Zhiwei Jiang +3 more
Diffusion large language models (dLLMs) enable flexible masked generation through bidirectional context modeling, but this capability also introduces...
Roberto Passaro, Ariel Flint, Romualdo Pastor-Satorras +2 more
As large language models (LLMs) are deployed in interacting populations, their collective behaviour depends not only on the models but also on who...
Wonjune Seo, Joonhyuk Baek, Yeseong Jung +1 more
Despite recent advances, deepfake detectors remain vulnerable to adversarial examples, particularly in diverse, real-world settings. We propose...
Mrityunjay Shukla, Shubham Roy, Biswabandan Panda
Large language models (LLMs) achieve high inference efficiency using optimized GPU kernels generated by just-in-time (JIT) compilation frameworks...
Michele Guerra, Wuyeh Jobe, Sanjana Nambiar +2 more
Web-searching Large Language Models (LLMs) have changed how users seek online information: instead of sifting through ranked lists of links, users...
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.
AI Threat Alert indexes 4,319+ papers on AI/ML security, classified across attack, defense, benchmark, survey, and tool categories and updated continuously.
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.
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.
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.
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
AI Threat Alert