Help Net Security

Prove has launched the Prove Identity Platform, turning identity verification into an ongoing, real-time process for users, businesses, and AI agents. AI agents are already initiating real transactions on behalf of real people. OpenAI and Stripe launched the Agentic Commerce Protocol in September. Visa named Anthropic, OpenAI, and Perplexity as agentic commerce partners. As that shift accelerates, the central question facing every organization becomes: Is this person real, and did they authorize this specific action, … More →

The post Prove Identity Platform connects verification, authentication, and fraud prevention appeared first on Help Net Security.

Hidden inside newly discovered botnet malware is an unusual message from its creator: “AI.NEEDS.TO.DIE”. Dubbed “tuxnokill” by researchers at Akamai, the malware is one of two fresh Mirai botnet variants documented this month by major cybersecurity firms and, judging by the aforementioned hard-coded string, this particular variant might have been coded the old-fashioned way. “Tuxnokill” and “Nexcorium” Based on hits on the company’s global network of honeypots, Akamai found that tuxnokill is spreading through CVE-2025-29635, … More →

The post New Mirai variants target routers and DVRs in parallel campaigns appeared first on Help Net Security.

Acronis has launced Acronis GenAI Protection, a monitoring and security solution that enables managed service providers (MSPs) to control generative AI usage across client environments, preventing sensitive data exposure and protecting against malicious prompt manipulation. Acronis GenAI Protection represents the initial phase of Acronis Cyber Workspace, with additional capabilities planned for release to deliver a protected AI workspace, natively integrated into the Acronis platform. As organisations rapidly adopt generative AI tools, businesses face growing risks … More →

The post Acronis GenAI Protection gives MSPs control over AI usage and data risks appeared first on Help Net Security.

Elastic has announced MCP Apps for Elastic, delivering agent-native UI experiences for security and observability workflows across third-party coding tools and chat clients. The new MCP Apps enable teams to investigate threats, diagnose system behavior, and act on data directly within the AI tools they already use, without switching tools or stitching together separate systems. Built on the Model Context Protocol (MCP) apps spec, the open standard co-authored by Anthropic and OpenAI, these apps allow … More →

The post Elastic MCP Apps bring security and observability workflows into AI tools appeared first on Help Net Security.

Progress Software has fixed a slew of high-severity vulnerabilities in MOVEit WAF and LoadMaster, including a flaw (CVE-2026-21876) that may allow attackers to bypass firewall detection. MOVEit WAF (web application firewall) is designed to protect Progress’s managed file transfer platform MOVEit Transfer from web-based attacks. (A zero-day vulnerability in MOVEit Transfer was infamously exploited in 2023 by the Cl0p cyber extortion gang to grab data from hundreds of organizations.) LoadMaster is the company’s general-purpose enterprise … More →

The post Progress Software fixes sneaky WAF bypass vulnerability (CVE-2026-21876) appeared first on Help Net Security.

Tencent has opened an international beta of QClaw, an AI agent application aimed at consumers in Canada, Japan, Singapore, South Korea, and the United States. The first wave is capped at 20,000 users. Additional markets are scheduled to follow. QClaw runs on Windows and MacOS. Setup takes about three minutes and involves downloading the application, registering an account, and scanning a QR code. The software is built on OpenClaw, the open-source personal AI assistant project. … More →

The post Tencent’s QClaw AI agent app arrives on Windows and macOS appeared first on Help Net Security.

Phishing returned as the leading method attackers used to break into organizations in the first quarter of 2026, accounting for over a third of engagements where initial access could be determined, according to Cisco Talos. It is the first quarter phishing has led the category since Q2 2025, when exploitation of public-facing applications took over following widespread attacks against on-premises Microsoft SharePoint servers. That SharePoint exploitation wave, collectively tracked as ToolShell, drove public-facing application exploitation … More →

The post Phishing reclaims the top initial access spot, attackers experiment with AI tools appeared first on Help Net Security.

Microsoft OneDrive’s recent updates focus on improving intelligence, collaboration, and administrative control. “Last year, we made a promise: your files should work for you, not the other way around. That meant reimagining OneDrive not just as a place to store files, but as an intelligent layer that surfaces what matters, eliminates busy work, and enables seamless collaboration across your organization. OneDrive became smarter, more collaborative, and more central to how work flows across Microsoft 365,” … More →

The post OneDrive updates focus on AI, access control, and compliance appeared first on Help Net Security.

Penetration testers have long relied on collections of specialized tools, manual coordination, and documented runbooks to work through a target assessment. PentAGI, an open-source project from VXControl, attempts to automate that entire workflow using a multi-agent AI system that plans, researches, and executes penetration tests with minimal human direction. How the agent system works PentAGI organizes work into a hierarchy of flows, tasks, subtasks, and actions. An orchestrator agent receives a goal and coordinates three … More →

The post PentAGI: Open-source autonomous AI penetration testing system appeared first on Help Net Security.

Apple claims that Apple Intelligence, a GenAI service provided on its operating systems, is designed with an extra focus on user security and privacy through a two-stage authentication and authorization system using anonymous access tokens. However, researchers from The Ohio State University have identified vulnerabilities in this design, demonstrated on macOS 26.0 (Tahoe), that allow attackers to steal and reuse these tokens. Service infrastructure The system offloads complex requests to cloud servers using Private Cloud … More →

The post Apple Intelligence flaw kept stolen tokens reusable on another device appeared first on Help Net Security.

Here’s an overview of some of last week’s most interesting news, articles, interviews and videos: Bringing governance and visibility to machine and AI identities In this Help Net Security interview, Archit Lohokare, CEO of AppViewX, explains how the rise of AI marked a turning point where machine and AI agent identities began converging into a single problem. Drawing on his experience across IBM and CyberArk, he describes the shift from human-driven systems to autonomous machines. … More →

The post Week in review: Acrobat Reader flaw exploited, Claude Mythos offensive capabilities and limits appeared first on Help Net Security.

Google claims that its security teams work around the clock using its Gemini AI models to detect and stop harmful ads. “Bad actors are using generative AI to create deceptive ads at scale, and Gemini helps us detect and block them in real time” Keerat Sharma, VP and GM, Ads Privacy and Safety, Google, said. “Our models analyze hundreds of billions of signals — including account age, behavioral cues and campaign patterns — to stop … More →

The post Google wipes out 602 million scam ads with Gemini on duty appeared first on Help Net Security.

The security researcher who earlier this month published a proof-of-concept (PoC) exploit for a zero-day privilege escalation vulnerability in Microsoft Defender is back with two more. The first, dubbed “RedSun,” is another privilege escalation flaw in the same platform. The second, “UnDefend,” allows a standard user to block Microsoft Defender from receiving signature updates or disable it entirely (if Microsoft pushes a major Defender update). And, according to Huntress researchers, all three exploitation techniques have … More →

The post Researcher drops two more Microsoft Defender zero-days, all three now exploited in the wild appeared first on Help Net Security.

GitLab has released GitLab 18.11, expanding agentic AI across the entire software lifecycle with security remediation, pipeline configuration, and delivery analytics. AI-generated code moves faster than the systems around it can keep up with, creating the AI paradox: faster code generation without faster delivery, security, or operations to match. As code volume grows, so does the backlog of pipelines to configure, security findings to remediate, and delivery questions to answer. GitLab 18.11 helps address those … More →

The post GitLab 18.11 brings agentic AI to security fixes, CI pipelines, and delivery analytics appeared first on Help Net Security.

Liongard has announced the expansion of LiongardIQ with new capabilities spanning programmatic AI integration, conversational querying, enhanced network discovery, and deeper identity mapping, extending its system of authority for asset intelligence across the full IT stack. Without a trusted foundation of asset intelligence, automation runs on assumptions, AI agents query stale data, and security gaps go undetected. LiongardIQ is that foundation, continuously discovering and monitoring every asset, identity, and configuration across the IT stack. “The … More →

The post Liongard upgrades LiongardIQ with AI access, live asset data, and deeper discovery appeared first on Help Net Security.

For organizations that want to keep company data within their own systems and have more control over how AI is deployed, Mozilla is offering an alternative to externally hosted AI services with Thunderbolt, an open-source AI client designed for self-hosted use. Thunderbolt is available through a waitlist, with native applications for web, macOS, Windows, Linux, iOS, and Android. Its source code is published on GitHub. Mozilla says Thunderbolt “gives organizations what proprietary AI services can’t,” … More →

The post Mozilla challenges enterprise AI providers with Thunderbolt, open-source AI client under your control appeared first on Help Net Security.

OpenAI is rolling out a major update to the Codex desktop app for users signed in with ChatGPT. Personalization features, including context-aware suggestions and memory, will roll out to Enterprise, Edu, and users in the EU and UK soon. Computer use is initially available on macOS and will expand to EU and UK users in the near future. Screenshot of Codex computer use on Mac (Source: OpenAI) The update expands Codex into a more capable … More →

The post Codex can now operate between apps. Where are the boundaries? appeared first on Help Net Security.

Google shipped Android 17 Beta 4 on April 16, marking the last scheduled beta in the Android 17 release cycle. The build targets app compatibility testing and platform stability ahead of the final release, and it carries several behavior changes that developers need to account for before the stable version ships. Supported Pixel devices can enroll in the Android Beta program to receive the update over the air. Developers without a Pixel device can use … More →

The post Android 17 Beta 4 arrives with post-quantum cryptography and new memory limits appeared first on Help Net Security.

Apple’s AirTag is designed to help users track lost items by relying on a vast network of nearby Apple devices. New research shows that this same system can be manipulated to display locations where an AirTag has never been. Relaying an AirTag’s BLE advertisments over the Internet injects false location reports into the Find My system How the attack works The Find My network depends on Bluetooth Low Energy (BLE) signals broadcast by AirTags. When … More →

The post Apple AirTag tracking can be misled by replayed Bluetooth signals appeared first on Help Net Security.

Governments are moving to block children under 16 from social media in the name of safety. But once these measures move from policy to practice, they raise a harder question: what happens when protecting kids requires collecting more data than ever before and may put them at greater risk? Age checks spark debate over privacy and safety Last year, Australia became the first country to introduce a ban for those under 16. Following its example, … More →

The post Social media bans might steer kids into riskier corners of the internet appeared first on Help Net Security.