关注Claude AI模型带来的金钱安全隐患，揭示AI如何放大网络攻击能力。
Article URL: https://www.rte.ie/brainstorm/2026/0517/1572343-claude-ai-model-security-issues-humans-work-information-data/ Comments URL: https://news.…
揭示大模型可能产生欺骗性推理链，对AI安全研究提出重要警示。
arXiv:2510.00319v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) have been demonstrating strong reasoning capability with their …
新研究揭示AI模型在武装冲突地区部署时，可能因对齐失败而无意中激化矛盾，对AI安全与全球治理敲响警钟。
arXiv:2605.22720v1 Announce Type: new Abstract: AI models are already deployed in societies affected by armed conflict, and journalists, humanitarian …
通过选择性几何控制，提升大模型安全对齐的鲁棒性，为AI防御攻击提供新思路。
arXiv:2602.07340v2 Announce Type: replace Abstract: Safety alignment of large language models remains brittle under domain shift and noisy preference …
研究发现多重共线性会严重削弱AI入侵检测的可解释性稳定性，并提出缓解策略，值得关注。
arXiv:2605.22529v1 Announce Type: cross Abstract: This paper investigates a unexplored yet impactful vulnerability in AI explainability used in intrus…
针对LLM在分布外场景下的对齐失败问题，提出基准测试和改进监测器的新方法，提升AI安全可靠性。
arXiv:2605.21602v1 Announce Type: new Abstract: Many safety and alignment failures of large language models (LLMs) occur due to out-of-distribution (O…
六大模型压力测试揭示AI安全护栏在持续追问下会失效，值得关注
This is a submission for the Google I/O Writing Challenge This is a submission for the Google I/O Writing Challenge We treat AI safety as a static sta…
用一致性训练算法对抗社交网络政治操纵，为AI安全提供新思路
arXiv:2605.22771v1 Announce Type: new Abstract: Large language models (LLMs) exhibit systematic political bias across a variety of sensitive contexts.…
提出Agentic CLEAR框架，自动化评估LLM Agents多层次能力，提升评估效率与客观性。
arXiv:2605.22608v1 Announce Type: new Abstract: Agentic systems are becoming more capable: agents define strategies, take actions, and interact with d…
被ICML 2026接收，研究如何通过高效学习构建AI合规护栏，兼顾安全与性能。
arXiv:2510.03485v2 Announce Type: replace Abstract: Autonomous web agents are increasingly deployed for long-horizon tasks, yet their ability to adher…
大模型会听指令还是学案例？论文揭示LLMs在指令服从与上下文归纳之间的行为冲突，揭秘“说一套做一套”的根源。
arXiv:2605.20382v1 Announce Type: new Abstract: Language models are trained to follow instructions, but they are also powerful pattern completers. Wha…
这篇 arXiv 论文探讨语言模型匿名化方法，为隐私保护提供新思路，适合 AI 安全研究者参考。
arXiv:2501.02407v3 Announce Type: replace-cross Abstract: Rapid advances in Natural Language Processing (NLP) have revolutionized many fields, includi…
针对LLM越狱新方法，采用自适应探针引导，克服了传统对比引导的偏差和手动调参局限，提升鲁棒性与有效性。
arXiv:2605.20286v1 Announce Type: cross Abstract: Recent work has demonstrated the potential of contrastive steering for jailbreaking Large Language M…
提出内部化逐步反思机制，让模型自主识别并防御间接越狱攻击，AI安全新范式。
arXiv:2605.20654v1 Announce Type: new Abstract: While Large Language Models (LLMs) demonstrate remarkable capabilities, they remain susceptible to sop…
首个专攻AI Agent监控器逃避攻击的基准测试，揭示安全漏洞新维度。
arXiv:2605.16626v2 Announce Type: replace-cross Abstract: Since autonomous coding agents generate complex behaviors at high-volume, we may want to use…
用草稿模型在推理前拦截有害输出，为AI安全提供轻量级新方案
arXiv:2605.19321v1 Announce Type: cross Abstract: Large Language Model (LLM) alignment remains vulnerable to jailbreak attacks that elicit unsafe resp…
多轮多模态攻击防不胜防？这篇论文提出预测性防御机制，应对未见过的新型攻击。
arXiv:2605.18988v1 Announce Type: cross Abstract: The expansion of Multimodal Large Language Models (MLLMs) and their integration into autonomous agen…
从内感受差异切入审美评价，揭示人类与AI对齐的新维度。
arXiv:2605.18759v1 Announce Type: cross Abstract: Artificial intelligence (AI), exemplified by large language models (LLMs), is rapidly approaching an…
用注意力机制优化强化学习奖励，成功攻破大型推理模型的安全防线，揭示AI安全新挑战。
arXiv:2605.19485v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) have demonstrated remarkable capabilities in solving complex problems by…
本地跑大模型做安全检查？用qwen-coder-32k在MacBook M5上实测，轻量高效还保护隐私。
Continuing articles AI-Powered Repository Security Check with Antigravity Workflow and https://dev.to/gdg/how-to-build-a-custom-ai-quality-gate-on-clo…