Falcon AIDR Detects Threats at the Prompt Layer in Kubernetes AI Applications
Falcon AIDR Detects Threats at the Prompt Layer in Kubernetes AI Applications
2026-05-26 归档
今日共收录 618 条安全情报，包含 0 个 CVE，271 篇安全通告，以及 95 篇研究论文。
← 返回归档列表Threat IntelligenceWelcome to BlackFile: Inside a Vishing Extortion OperationBy Google Threat Intelligence Group • 16-minute read
Threat IntelligenceWelcome to BlackFile: Inside a Vishing Extortion OperationBy Google Threat Intelligence Group • 16-m
Now Live: The CrowdStrike 2026 Financial Services Threat Landscape Report
Now Live: The CrowdStrike 2026 Financial Services Threat Landscape Report
How to Protect Identities and Sessions from Infostealers
How to Protect Identities and Sessions from Infostealers
New Claude Integration Brings Audit Data into the Falcon Platform
New Claude Integration Brings Audit Data into the Falcon Platform
Measuring AI-Enabled Success: 3 KPIs Leaders Should TrackMay 22, 2026
Measuring AI-Enabled Success: 3 KPIs Leaders Should TrackMay 22, 2026
Threat Intelligence2 PhaaS 2 Furious: The Evolution of Chinese-language Phishing ServicesBy Google Threat Intelligence Group • 7-minute read
Threat Intelligence2 PhaaS 2 Furious: The Evolution of Chinese-language Phishing ServicesBy Google Threat Intelligence G
TeamPCP Supply Chain Campaign: Activity Through 2026-05-24, (Mon, May 25th)
TeamPCP now operates across three package ecosystems in parallel, it reached GitHub&#;x26;#;39;s own internal codebase,
TeamPCP Supply Chain Campaign: Activity Through 2026-05-24, (Mon, May 25th)
TeamPCP now operates across three package ecosystems in parallel, it reached GitHub&#;x26;#;39;s own internal codebase,
Anthropic’s restricted Claude Mythos model may be coming to Claude Code
Anthropic appears to be preparing for the public rollout of the Mythos model, which was announced in April as a restrict
FBI warns of Kali365 phishing service targeting Microsoft 365 accounts
The FBI is warning about the Kali365 phishing-as-a-service platform (PhaaS) that is used to hijack Microsoft 365 account
Lazarus Deploys RemotePE Memory-Only RAT Against Financial and Crypto Firms
Cybersecurity researchers have shed light on a cross-platform malware called RemotePE that has been put to use by the No
TrapDoor Supply Chain Attack Spreads Credential-Stealing Malware via npm, PyPI, and CratesIO
A new coordinated cross-ecosystem software supply chain attack campaign has targeted npm, PyPI, and Crates.io to distrib
266,000 Affected by Data Breach at Radiology Associates of Richmond
Threat actors stole files containing names and protected health information from the healthcare organization’s systems.
Anthropic: Mythos Detected 23,000 Potential Vulnerabilities Across 1,000 OSS Projects
Many findings have been confirmed to be critical or high-severity vulnerabilities and the number will continue to increa
Over 5,500 GitHub Repositories Infected in ‘Megalodon’ Supply Chain Attack
Fake automated commits injected GitHub Actions workflows containing payloads to steal credentials, CI secrets, keys, and
Five Queries Are Enough: Query-Efficient and Surrogate-Free Membership Inference Attacks on RAG via Entailment
本文研究检索增强生成（RAG）系统中的成员推断攻击（MIA）问题。RAG通过将外部知识库引入LLM响应，有效减少幻觉，但同时也带来了新的隐私风险：攻击者可能通过模型输出推断检索语料库中是否存在特定文档，从而泄露敏感信息。现有MIA方法要么依
Ellipsoid Control: A White-list Jailbreak Defense via Benign Latent Modeling
该论文针对大型语言模型（LLM）的越狱攻击防御问题，指出现有的表示工程（RepE）防御方法本质上是基于黑名单监督的：它们从有害或越狱数据中学习将越狱表示转换为拒绝表示的激活变换，但这类数据本身不完整且不断演化，导致防御性能严重依赖收集样本的
Poisoning the Watchtower: Prompt Injection Attacks Against LLM-Augmented Security Operations Through Adversarial Log Content
该论文研究了大型语言模型（LLM）作为安全运营中心（SOC）分析师辅助工具时的结构性安全缺陷。在SOC中，LLM通常被用于处理日志和告警数据，以生成分流标签、事件摘要或补救建议。然而，许多日志字段（如用户代理、URL、payload、DNS
Broken Object Level Authorization in the Wild: An Empirical Taxonomy from 100+ Bug Bounty Disclosures
对象级授权失效（BOLA）是OWASP API安全漏洞列表中长期排名最高的威胁，但此前研究多停留在概念层面。本文基于公开的漏洞赏金报告，首次对BOLA进行大规模实证分析。研究者从HackerOne平台筛选出2021-2026年间标记为IDO