Aggregator

抢先加入AI时代顶尖安全团队！阿里云2027届实习生招聘来了！

Iran’s “hacktivist” group Ababil of Minab, which hit LA Metro and wiped terabytes of data, is forensically linked to Iran’s intelligence service MOIS. In late March, a group calling itself Ababil of Minab posted videos and screenshots online claiming it had broken into the Los Angeles County Metropolitan Transportation Authority, wiped hundreds of terabytes of […]

Как Huawei собирается резко увеличить плотность транзисторов без уменьшения техпроцесса.

Anthropic introduced a security-guidance plugin for Claude Code that reviews code changes for common vulnerabilities and helps Claude identify and fix issues during the same development session. The company says the plugin is designed to catch issues such as injection flaws, unsafe deserialization, and insecure DOM APIs before code reaches pull requests, reducing the amount of manual security review later in the development process. Once installed, the plugin runs automatically during development sessions, without requiring … More →

The post Claude now reviews and fixes vulnerabilities as you write code appeared first on Help Net Security.

Cogent has launched two new platform capabilities designed to reduce the time between vulnerability disclosure and confirmed remediation. Zero Day Response identifies exposure within minutes of public disclosure, without waiting for scanner signatures. Autonomous Remediation determines the right fix, assesses business impact before execution, and confirms that the vulnerability has been resolved. The releases arrive as AI-assisted exploit development compresses attacker timelines faster than most security programs can keep pace. Time to exploit has collapsed … More →

The post Cogent targets exploit-to-remediation gap with new AI-powered security capabilities appeared first on Help Net Security.

CrowdStrike has dismantled the Glassworm botnet in an operation aided by Google and Shadowserver, stripping the operators’ access to infrastructure that helped threat actors infect hundreds of pieces of open-source software with malware since early 2025, the company said Tuesday.  The coordinated effort involved the simultaneous takedown of four attacker-controlled servers that were designed to […]

The post CrowdStrike disrupts Glassworm botnet that preyed on open-source supply chain appeared first on CyberScoop.

Cybermindz warns that cybersecurity burnout is a growing risk, urging organizations to move beyond wellness initiatives and adopt a measurable, risk-based approach to workforce stress

When an employee installs an AI writing assistant, connects a coding copilot to their IDE, or starts summarizing meetings with a new browser tool, they are doing exactly what a productive employee should do: finding faster ways to work. Across most organizations today, employees are running three to five AI tools on any given day. Most were never reviewed by IT. A significant portion connects

The Glassworm botnet targeting developers in software supply-chain attacks has been disrupted after researchers took down its resilient command-and-control infrastructure relying on Solana blockchain transactions and the BitTorrent DHT network. [...]

The suspect was detained in the central Dutch town of Buren, where law enforcement officers also searched his home and seized multiple digital storage devices, according to a statement released Tuesday by the Dutch National Police.

Ping Identity announced new capabilities that extend the Ping Identity Platform for the agentic enterprise, where AI agents, automation, and developers increasingly shape how access is managed, governed, and secured across organizations. AI agents are changing both sides of the identity equation. They are new actors that need to be discovered, governed, and managed across their lifecycle, and they are also new operators that can help builders administer and secure identity environments through machine-native interfaces. … More →

The post Ping Identity advances agentic security with AI governance and trusted access appeared first on Help Net Security.

恭喜系统软件与安全实验室2026届毕业生完成答辩，顺利毕业！

本文详细评估了阿里 AAIG 开源的独立大模型安全围栏 YuFeng-XGuard 的防御能力、核心优势以及现存的软肋

一次简单的小实验，文末有项目附件，欢迎交流！

The hacking group claimed to be a standalone hacktivist crew but actually has ties to the Ministry of Intelligence of the Islamic Republic of Iran (MOIS), researchers at Gambit Security said in a report published Tuesday.

在遇到vue框架的时候,使用相关插件进行接口相关的测试，往往更容易找到突破口。

Архитектурный изъян популярного Python-фреймворка Starlette превращает любую проверку доступа в иллюзию.

The old way used to be all about observability, dashboards, aggregated KPIs, human correlation, and manual intervention. That world is changing with AI. — Donogh O’Reilly, Vice President of Sales for NETSCOUT Artificial intelligence (AI) no longer sits on the fringe of the service provider’s transformation strategy. It...

这篇文章用 PB&J 三明治练习解释 Agent 安全：提示词注入之外，更棘手的是有授权工具的 Agent 按字面执行模糊指令，造成事故。

eSentire has unveiled new preempt, detect, and respond capabilities within the Atlas Platform, a unified agentic AI platform with purpose-built AI Operatives that work together in a continuous security lifecycle. Controlled autonomy SecOps The Atlas Platform delivers purpose-built and adaptive AI operative infrastructure in a continuous closed loop across autonomous AI offensive security, exposure management, and Managed Detection and Response (MDR) services. It executes hundreds of thousands of autonomous investigations and responses across 2,000+ customer … More →

The post eSentire launches new Atlas AI Operatives for autonomous threat detection and response appeared first on Help Net Security.