Cybersecurity Brief

Cloud Upgrades Lead; Rowhammer Bypass and Targeted Campaigns

Coverage: 15 Sept 2025 (UTC)

Cloud providers emphasized performance and operational efficiency today. AWS made new memory- and network-optimized EC2 R8gn instances generally available, while Azure Container Storage released a major update designed to accelerate AI, databases, and stateful Kubernetes workloads. Alongside platform changes, vendors highlighted detection improvements and new integrations, and researchers detailed a Rowhammer technique that defeats DDR5 mitigations, underscoring ongoing hardware risk.

Platform performance and AI upgrades

AWS introduced EC2 R8gn, a memory- and network-optimized instance family powered by Graviton4 and 6th-generation Nitro Cards. The lineup scales to 48xlarge with up to 1,536 GiB of memory and up to 600 Gbps of network bandwidth, with Elastic Fabric Adapter on larger sizes. The positioning targets network-intensive, latency-sensitive applications such as databases and distributed systems; customers should validate sizing, EBS and network profiles, and cluster tuning to capture expected gains.

Microsoft unveiled Azure Container Storage v2.0.0, engineered for local NVMe on select VM families to deliver roughly 7× higher IOPS and ~4× lower latency versus the prior release. The update removes minimum cluster size requirements, introduces an open-source local CSI driver, and eliminates per‑GB fees for larger pools, broadening use across production, dev/test, and edge scenarios. The release aims to speed AI inference and database performance, simplify operations, and expand deployment flexibility.

For teams operationalizing custom models, Amazon Bedrock added on-demand deployment for customized Meta Llama models, letting customers serve real-time inference without pre‑provisioned endpoints. The pay‑as‑you‑go option can lower costs for intermittent or variable traffic, with the usual trade‑offs around potential cold-start latency on highly latency‑sensitive workloads.

Google released an externalized build of its internal ML profiler as XProf, integrating into OpenXLA and supporting XLA-based frameworks such as JAX, PyTorch/XLA and TensorFlow/Keras. The Google Cloud update brings a trace viewer, memory viewer across multiple memory types, graph/HLO op profiling, roofline analysis, and faster loading of large profiles via a Cloud Diagnostics library that automates dependency packaging and TensorBoard provisioning.

OpenSearch 3.1 is now supported in Amazon OpenSearch Service, incorporating Lucene 10 for optimized vector indexing, memory‑mapped Faiss for efficiency, and improvements to range queries and high‑cardinality aggregations. New relevance tooling and hybrid search scoring refinements target both generative AI applications and traditional analytics.

On the desktop, Microsoft plans to automatically deploy the Microsoft 365 Copilot app to Windows devices with Microsoft 365 desktop clients starting in October and completing by mid‑November 2025 (EEA excluded). The change, communicated via a message center update, centralizes access to Copilot experiences and can be opted out at the tenant level. Organizations should prepare helpdesk communications, review deployment policies, and test in rings to limit user disruption. Details are in BleepingComputer.

Detection and response integrations

AWS outlined how GuardDuty protection plans and Extended Threat Detection work together to correlate telemetry from sources such as S3, EKS, runtime monitoring, RDS login profiling, and Lambda network activity. Extended Threat Detection is enabled by default, applies AI/ML to combine multi‑step signals into prioritized findings mapped to MITRE ATT&CK, and includes scenarios that show multi‑vector attack surfacing. A 30‑day trial per plan per account/region supports coverage evaluation.

Cloudflare and CrowdStrike connected Cloudflare One with Falcon Fusion SOAR to automate email and access containment and endpoint isolation. Out‑of‑the‑box connectors and prebuilt actions in Falcon enable playbooks that revoke tokens, update allow/block lists, adjust access policies, and isolate hosts based on detections, with bidirectional triggers via APIs and Logpush to accelerate SOC workflows.

Hardware and software weaknesses surface

Google detailed its support for DDR5 Rowhammer research, open‑sourcing FPGA-based test platforms and contributing to standardization efforts such as PRAC for deterministic row‑activation counting. In collaboration with ETH Zurich, researchers demonstrated an access pattern that bypasses enhanced Target Row Refresh on some DDR5 modules and achieved a privilege‑escalation exploit on a commodity system under lab conditions. Complementing this, BleepingComputer reported technical details of a new technique dubbed Phoenix that synchronizes with refresh behavior to induce bit flips across tested DDR5 chips, highlighting limits of probabilistic mitigations and the need for coordinated, deterministic defenses.

On mobile, CSOonline reported a critical vulnerability in Samsung’s Quramsoft image library that enables remote code execution when crafted images are processed and has been exploited in the wild. Samsung issued a September security update; organizations should prioritize patching managed and BYOD fleets and monitor for anomalous device activity. In manufacturing IT, CSOonline covered active exploitation of a deserialization flaw in Delmia Apriso across multiple releases, with calls to expedite updates and apply compensating controls due to operational complexity.

Incidents and targeted campaigns

BleepingComputer reported that FinWise Bank disclosed an insider incident affecting American First Finance customers after a former employee accessed systems post‑termination, with 689,000 individuals noted in a partner notification. The bank engaged external investigators and is offering credit monitoring. Separately, BleepingComputer covered Google’s confirmation that a fraudulent account briefly accessed its Law Enforcement Request System portal before being disabled; the company said no requests were submitted and no data was accessed.

The Hacker News highlighted IBM X‑Force research attributing a Thailand‑focused campaign to Mustang Panda, including a USB worm (SnakeDisk) that drops the Yokai backdoor and geofences execution to Thai IP addresses. In broader commodity malware operations targeting Chinese‑speaking users, The Hacker News and partners documented SEO poisoning and abuse of legitimate hosting to deliver RAT families such as HiddenGh0st, Winos, kkRAT, and FatalRAT via trojanized installers and DLL sideloading.

On credential theft, CSOonline covered Okta’s analysis of a PhaaS framework dubbed VoidProxy that runs adversary‑in‑the‑middle phishing against Microsoft, Google, and multiple SSO providers to capture credentials, MFA responses, and session cookies. Recommended defenses include phishing‑resistant authentication, risk‑based access controls, and automated containment for suspicious sessions.

These and other news items from the day:

Mon, September 15, 2025

AWS Launches General Availability of EC2 R8gn Instances

🚀 AWS has made the new Amazon EC2 R8gn instances generally available, built on AWS Graviton4 processors to deliver up to 30% better compute performance than Graviton3. R8gn instances include 6th generation Nitro Cards and offer up to 600 Gbps network bandwidth—the highest among network-optimized EC2 instances—and sizes up to 48xlarge with metal options. They provide up to 1,536 GiB memory, up to 60 Gbps EBS bandwidth, and support Elastic Fabric Adapter (EFA) on select large and metal sizes for lower latency cluster performance.

read more →

Mon, September 15, 2025

On-demand deployment for custom Meta Llama models on Bedrock

🚀 Amazon Bedrock now offers an on-demand deployment option for customized Meta Llama 3.3 models that have been fine-tuned or distilled in Bedrock; models customized on or after September 15, 2025 are eligible. The feature lets customers process requests in real time and pay only for consumed compute, removing the need for pre-provisioned always-on resources. Bedrock continues to provide a managed platform with built-in security, privacy, and responsible AI capabilities.

read more →

Mon, September 15, 2025

Google releases XProf and Cloud Diagnostics XProf tools

🔧 Google has open-sourced XProf, an upgraded ML profiler, and published the Cloud Diagnostics XProf library to simplify profiling and optimizing models on xPUs. The release brings unified XLA-based profiling across JAX, PyTorch/XLA and TensorFlow/Keras, and supports programmatic and on-demand trace capture. The Cloud Diagnostics library packages dependencies, stores profiles in Google Cloud Storage for retention, provisions TensorBoard on VMs or GKE for faster loading, and produces shareable links for collaborative analysis with tunable machine types for performance.

read more →

Mon, September 15, 2025

Amazon OpenSearch Service Adds AI-Powered Forecasting

📈 Amazon now lets customers generate AI-powered forecasts and visualizations from time-series data indexed into Amazon OpenSearch Service domains. Forecasts can enhance insights into infrastructure utilization, application and business metrics, website traffic, and system performance to help anticipate upcoming changes. You can set up forecasts directly from OpenSearch dashboards or the OpenSearch UI, and no data science or AI expertise is required. The feature is available in all regions that support OpenSearch 3.1+ domains.

read more →

Mon, September 15, 2025

Microsoft to Auto-Install 365 Copilot on Windows in Oct

📥 Microsoft will automatically install the Microsoft 365 Copilot app on Windows devices that already have Microsoft 365 desktop apps beginning in early October, with rollout completing by mid-November 2025. The app will be placed in the Start menu and enabled by default, but administrators can opt out via the Apps Admin Center. Systems in the EEA are excluded, and Microsoft advises IT to notify helpdesk teams and users beforehand to reduce confusion and support requests.

read more →

Mon, September 15, 2025

Azure Container Storage v2.0.0: NVMe Boosts Kubernetes

⚡ Azure today released Azure Container Storage v2.0.0, a performance-first update that delivers up to 7× higher IOPS, 4× lower latency, and improved resource efficiency for Kubernetes stateful workloads. The release adds built-in support for local NVMe drives, removes prior pricing tiers for large pools, and is available as an open-source local CSI driver for non-AKS clusters. Optimized for storage- and GPU-optimized VM families, the update also enables single-node deployments and integrates with KAITO to speed AI model loading and scaling.

read more →

Mon, September 15, 2025

Amazon GuardDuty Protection Plans and Threat Detection

🔐 Amazon GuardDuty centralizes continuous threat detection across AWS using AI/ML and integrated threat intelligence. It offers optional protection plans—S3, EKS, Runtime Monitoring, Malware Protection for EC2 and S3, RDS, and Lambda—that extend detections to service-specific telemetry and runtime behaviors. Built-in Extended Threat Detection correlates signals into high-confidence attack sequences and maps findings to MITRE ATT&CK, providing prioritized remediation guidance.

read more →

Mon, September 15, 2025

Amazon OpenSearch Service adds OpenSearch 3.1 for vectors

🚀 Amazon OpenSearch Service now supports OpenSearch 3.1, bringing targeted improvements for vector-driven and traditional search workloads. The release bundles Lucene 10 for optimized vector field indexing, faster indexing times, reduced index sizes, sparse indexing, and vector quantization to lower memory usage. It also improves range query and high-cardinality aggregation latency and introduces a new Search Relevance Workbench for iterative quality testing. Additional vector search enhancements include Z-score normalization for more reliable hybrid search and memory-optimized Faiss support; OpenSearch 3.1 is available in all AWS Regions.

read more →

Mon, September 15, 2025

Critical RCE in Delmia Apriso Triggers Urgent Patching

⚠ A critical remote code execution flaw, CVE-2025-5086, has been observed being exploited in the wild against Delmia Apriso, Dassault Systèmes' manufacturing operations platform. CISA added the issue to its Known Exploited Vulnerabilities catalog with a CVSS score of 9.0, yet the vendor has provided minimal public guidance. Researchers report exploit scans and a circulating sample that was detected by only one AV engine, underscoring urgent patching challenges for manufacturers.

read more →

Mon, September 15, 2025

Phoenix Rowhammer Bypass Targets DDR5 TRR Defenses

🧨 Researchers have developed Phoenix, a new Rowhammer variant that defeats DDR5 TRR protections on SK Hynix modules by synchronizing and self-correcting against missed refresh intervals. After reverse-engineering TRR behavior, the team identified refresh slots that were not sampled and used precise hammering patterns covering 128 and 2,608 refresh intervals to flip bits. In tests they flipped bits across all tested DIMMs and produced a working privilege-escalation exploit, achieving a root shell on commodity DDR5 systems in under two minutes. The authors published an academic paper and an FPGA-based repository with experiments and proof-of-concept code.

read more →

Mon, September 15, 2025

Supporting Rowhammer Research to Strengthen DDR5 Mitigations

🔬 Google funded and collaborated on open-source DDR5 Rowhammer test platforms and academic research to evaluate current in-DRAM mitigations. Working with Antmicro and ETH Zurich, the team produced FPGA-based RDIMM and SO‑DIMM testers and used them to discover the Phoenix attack family, which includes a self-correcting refresh synchronization technique that can bypass enhanced TRR on some DDR5 modules. Google also led JEDEC standardization work on PRAC to enable deterministic row-activation counting and continues to share tools and findings to improve defenses.

read more →

Mon, September 15, 2025

Samsung image library flaw enables zero-click RCE exploit

📸 Samsung disclosed a critical remote code execution vulnerability in a closed-source image-parsing library, libimagecodec.quram.so, supplied by Quramsoft that affects devices running Android 13–16. The out-of-bounds write (CVE-2025-21043, CVSS 8.8) can be triggered by a specially crafted image and has been exploited in the wild. Messaging apps are a likely vector and the flaw can operate as a zero-click backdoor. Samsung released an SMR Sep-2025 Release 1 patch; enterprises should prioritize deployment.

read more →

Mon, September 15, 2025

Code Assistant Risks: Indirect Prompt Injection and Misuse

🛡️ Unit 42 describes how IDE-integrated AI code assistants can be abused to insert backdoors, leak secrets, or produce harmful output by exploiting features like chat, auto-complete, and context attachment. The report highlights an indirect prompt injection vector where attackers contaminate public or third‑party data sources; when that data is attached as context, malicious instructions can hijack the assistant. It recommends reviewing generated code, controlling attached context, adopting standard LLM security practices, and contacting Unit 42 if compromise is suspected.

read more →

Mon, September 15, 2025

FinWise Insider Data Breach Affects 689K AFF Customers

🔒 FinWise Bank says a former employee accessed sensitive files after their employment ended, in a data security incident identified on May 31, 2024. The bank notified corporate partner American First Finance (AFF), which reported that data for 689,000 customers was affected. FinWise launched an external investigation, strengthened internal controls, and is offering 12 months of credit monitoring and identity theft protection to impacted individuals.

read more →

Mon, September 15, 2025

VoidProxy Phishing Framework Bypasses MFA for SSO Logins

🔒 Okta threat researchers have identified a Phishing-as-a-Service called VoidProxy that leverages Adversary-in-the-Middle techniques to capture usernames, passwords, MFA codes and session cookies from Microsoft, Google and several SSO providers. The service uses compromised email service provider accounts, URL shorteners, Cloudflare Workers and disposable domains to evade detection and takedown. Victim credentials and session tokens are proxied to legitimate services, allowing attackers to reuse valid session cookies. Okta warns legacy methods such as SMS and OTP are especially vulnerable to this attack.

read more →

Mon, September 15, 2025

Kimsuky Uses AI to Forge South Korean Military ID Images

🛡️Researchers at Genians say North Korea’s Kimsuky group used ChatGPT to generate fake South Korean military ID images as part of a targeted spear-phishing campaign aimed at inducing victims to click a malicious link. The emails impersonated a defense-related institution and attached PNG samples later identified as deepfakes with a 98% probability. A bundled file, LhUdPC3G.bat, executed malware that enabled data theft and remote control. Primary targets included researchers, human-rights activists and journalists focused on North Korea.

read more →

Mon, September 15, 2025

HiddenGh0st, Winos and kkRAT Abuse SEO and GitHub Pages

🚨 Fortinet and Zscaler researchers describe an SEO poisoning campaign that targets Chinese-speaking users by surfacing spoofed download pages and GitHub Pages that host trojanized installers. Attackers manipulated search rankings and registered lookalike domains to trick victims into downloading installers bundling legitimate applications with hidden malware such as HiddenGh0st and Winos. Delivery chains use scripts (for example, nice.js), multi-stage JSON redirects, malicious DLLs and DLL sideloading to evade detection and establish persistence.

read more →

Mon, September 15, 2025

Cloudflare integrates CrowdStrike Falcon Fusion SOAR

🔗 Cloudflare announced an integration between the Cloudflare One SASE platform and CrowdStrike Falcon Fusion SOAR, delivering two out‑of‑the‑box connectors for Zero Trust and Email Security. The prebuilt actions exposed in the CrowdStrike Content Library automate common tasks—searching messages, updating allow/block lists, adjusting access policies, and revoking tokens—to reduce manual investigation and accelerate remediation. Customers can chain Cloudflare actions with Falcon Fusion playbooks via a drag‑and‑drop editor to enable bidirectional containment across network, email, and endpoints. The integration supports Logpush to CrowdStrike HTTP ingest and can be enabled from both vendor consoles, with APIs and custom playbooks available for tailoring workflows.

read more →

Mon, September 15, 2025

Google: Fraudulent Account Created in Law Enforcement Portal

🔒 Google confirmed that a fraudulent account was created in its Law Enforcement Request System (LERS) portal and has been disabled. The company said no requests were made with the account and no data was accessed. The claim follows posts by a group calling itself "Scattered Lapsus$ Hunters", which also asserted access to the FBI's eCheck system. The actors have previously targeted Salesforce-related infrastructure and taunted security teams.

read more →

Mon, September 15, 2025

Mustang Panda Uses SnakeDisk USB Worm to Deliver Yokai

🐍 IBM X-Force reports that China-aligned Mustang Panda is deploying a new USB worm, SnakeDisk, to propagate the Yokai backdoor against machines geolocated to Thailand. The actor also introduced updated TONESHELL variants (TONESHELL8/9) with proxy-aware C2 and parallel reverse shells. SnakeDisk abuses DLL side-loading and USB volume masquerading—moving user files into a subfolder and presenting a deceptive 'USB.exe' lure before restoring originals—to spread selectively on Thailand-based public IPs.

read more →

Mon, September 15, 2025

SEO Poisoning Targets Chinese Windows Users at Scale

🔍 Security researchers at FortiGuard Labs uncovered an SEO poisoning campaign that manipulated search results to steer Chinese-speaking Microsoft Windows users to spoofed download sites. Attackers registered lookalike domains and used subtle character substitutions to present compromised installers that bundled legitimate apps with hidden malware such as Hiddengh0st and Winos. The operation used a redirection script known as nice.js, anti-analysis checks in components like EnumW.dll, and persistence mechanisms including registry changes and TypeLib hijacking. FortiGuard warns the final payloads supported monitoring, keystroke and clipboard capture, Telegram interception, and cryptocurrency wallet theft.

read more →

Mon, September 15, 2025

Experts Say Scattered Spider 'Retirement' Is a Smokescreen

🕵️ Scattered Spider and roughly 15 affiliated ransomware and cybercrime groups posted a joint manifesto on BreachForums claiming to 'go dark' after recent arrests. Experts point to inconsistencies — an unlikely coalition, rapid timing, and no observed money‑movement — and call the announcement a likely smokescreen. They warn organizations not to lower their guard and to assume tactics and infrastructure remain active, taking immediate hardening steps.

read more →

Mon, September 15, 2025

Yurei Ransomware Uses Open-Source Tools for Extortion

🔒 A newly identified ransomware group called Yurei is conducting double-extortion attacks, encrypting files and exfiltrating sensitive data before demanding payment. First observed by Check Point Research on September 5, Yurei has targeted organizations in Sri Lanka, India and Nigeria and may have ties to Morocco. Built largely from open-source Prince-Ransomware code, the malware encrypts each file using per-file ChaCha20 keys protected with ECIES, appending a .Yurei extension, and attempts to provide a ransom page and .onion contact. Although the early variant omits some operational features (for example it fails to set a ransom wallpaper and does not remove Windows shadow copies), the group still threatens publication of stolen data to pressure victims.

read more →

Mon, September 15, 2025

Whistleblower Lawsuit Alleges WhatsApp Security Failures

🛡️ Attaullah Baig, former head of security at WhatsApp, has filed a whistleblower lawsuit alleging that Facebook knowingly failed to fix multiple security flaws in breach of its 2019 settlement with the FTC. The complaint asserts that in 2022 roughly 100,000 accounts were compromised daily, rising to as many as 400,000 daily lockouts by last year, and that inadequate anti-scraping protections exposed profile data at scale. Baig invokes the whistleblower-protection provisions of the Sarbanes-Oxley Act, and the filing has prompted wider media coverage and potential regulatory scrutiny.

read more →

Mon, September 15, 2025

Microsoft: Exchange Server 2016/2019 End Support Soon

⚠️ Microsoft has warned that Exchange Server 2016 and Exchange Server 2019 will reach end of extended support on October 14, 2025. After that date Microsoft will stop providing technical support, including bug fixes, time zone updates, and security patches, which could increase exposure to vulnerabilities. Administrators are advised to migrate to Exchange Online or upgrade to Exchange Server Subscription Edition, with documented migration and upgrade paths available.

read more →

Mon, September 15, 2025

Amazon S3 Batch Operations: Manage Objects by Bucket

📦 Amazon S3 Batch Operations now allows users to target all objects within an S3 bucket, or refine selection by prefix, suffix, creation date, or storage class directly from the AWS Management Console. Instead of providing a manifest, customers can specify these filters when creating a job and S3 will apply the chosen operation to every matching object and produce a detailed completion report. Typical use cases include copying between buckets, restoring archived data from Glacier classes, and computing checksums for datasets. The capability is available in all AWS Regions and can also be accessed via the AWS CLI or SDK.

read more →

Mon, September 15, 2025

Amazon SageMaker HyperPod: Slurm Health Agent Now GA

🩺 Amazon announces general availability of the SageMaker HyperPod health monitoring agent for Slurm clusters. The agent runs continuously on GPU- and Trainium-based nodes to perform passive background checks, detect hardware faults (for example, unresponsive GPUs and NVLink errors), and mark and replace unhealthy nodes automatically. It supports automatic reboots and coordinates with Slurm job auto-resume so training can continue from the last checkpoint, reducing manual intervention and downtime.

read more →

Mon, September 15, 2025

APAC Security Leaders on AI: CISO Community Takeaways

🤖 At the Google Cloud CISO Community event in Singapore, APAC security leaders highlighted accelerating investment in cybersecurity AI to scale operations and enable business outcomes. They emphasized priorities: getting AI implementation and governance right, securing the AI supply chain, and translating cyber risk into board-level impact. Practical wins noted include reduced investigation time, agentic SOC automation, and strengthened threat intelligence sharing.

read more →

Mon, September 15, 2025

Why Hybrid Mesh Firewalls Are the Future — Fortinet

🛡️ Fortinet argues that hybrid mesh firewalls (HMF) are now the baseline for modern enterprise security, unifying hardware, virtual, cloud-native and FWaaS instances under a centralized management plane. The post highlights FortiGate appliances running a single OS (FortiOS) and custom ASICs as the foundation for consistent policy, high-performance inspection, and fabric-wide telemetry. It frames HMF as essential for SASE integration and scalable, consistent enforcement across clouds, branches and data centers.

read more →

Mon, September 15, 2025

AWS Organizations adds Account State field for members

🛈 AWS Organizations introduces a new State field in the console and APIs (DescribeAccount, ListAccounts, ListAccountsForParent) to provide more granular lifecycle visibility for member accounts. The console Status field has been replaced by State, while both Status and State remain available in APIs until September 9, 2026. New state values include SUSPENDED, PENDING_CLOSURE, and CLOSED (within the 90-day reinstatement window). Customers should update account vending pipelines and governance integrations to reference State before the Status deprecation date.

read more →

Mon, September 15, 2025

Weekly Recap: Bootkit Malware, AI Attacks, Supply Chain

⚡ This weekly recap synthesizes critical cyber events and trends, highlighting a new bootkit, AI-enhanced attack tooling, and persistent supply-chain intrusions. HybridPetya samples demonstrate techniques to bypass UEFI Secure Boot, enabling bootkit persistence that can evade AV and survive OS reinstalls. The briefing also covers vendor emergency patches, novel Android RATs, fileless frameworks, and practical patch priorities for defenders.

read more →

Mon, September 15, 2025

Phishing Campaigns Deploy RMM Tools via Multiple Lures

🔒 New phishing campaigns are delivering remote monitoring and management (RMM) software by using multiple realistic lures, security firms warn. Attackers spoof browser updates, meeting software installers, party e-invites and government forms to trick victims into running installers for ITarian (Comodo), Atera, PDQ, SimpleHelp and ScreenConnect. Some campaigns host payloads on trusted services such as Cloudflare R2 and may install multiple RMM tools in quick succession. Analysts caution RMM compromise can lead to ransomware and data theft and recommend endpoint detection, approved-tool enforcement and enhanced network controls such as browser isolation.

read more →

Mon, September 15, 2025

Microsoft removes upgrade block for Windows 11 audio

🔧 Microsoft has removed a safeguard hold that blocked upgrades to Windows 11 24H2 on devices running Dirac audio enhancement software after reports that the component cridspapo.dll caused integrated speakers and Bluetooth audio devices to stop working. A new driver is available via Windows Update and Microsoft recommends installing the latest security update; restarting the device may speed the offering. The safeguard hold was lifted on September 11, 2025, but other upgrade blocks remain for unrelated driver and software incompatibilities.

read more →

Mon, September 15, 2025

HybridPetya Resembles NotPetya and Adds UEFI Bootkit

🔒 ESET Research identified HybridPetya on VirusTotal in February 2025, with filenames implying a connection to the destructive NotPetya outbreak. The strain encrypts the NTFS Master File Table using Salsa20 and deploys a UEFI bootkit on the EFI System Partition to ensure firmware‑level persistence. One variant exploits CVE-2024-7344 to bypass UEFI Secure Boot via a signed but vulnerable Microsoft component, yet retains a working decryption mechanism for victims. Analysts found no signs of self-propagation like NotPetya, but the combination of pre-boot compromise and MFT encryption raises significant concern.

read more →

Mon, September 15, 2025

AI-Powered Villager Pen Testing Tool Raises Abuse Concerns

⚠️ The AI-driven penetration testing framework Villager, attributed to China-linked developer Cyberspike, has attracted nearly 11,000 PyPI downloads since its July 2025 upload, prompting warnings about potential abuse. Marketed as a red‑teaming automation platform, it integrates Kali toolsets, LangChain, and AI models to convert natural‑language commands into technical actions and orchestrate tests. Researchers found built‑in plugins resembling remote access tools and known hacktools, and note Villager’s use of ephemeral Kali containers, randomized ports, and an AI task layer that together lower the bar for misuse and complicate detection and attribution.

read more →

Mon, September 15, 2025

Five Trends Reshaping IT Security Strategies in 2025

🔒 Cybersecurity leaders report the mission to defend organizations is unchanged, but threats, technology and operating pressures are evolving rapidly. Five trends — shrinking or stagnating budgets, AI-enabled attacks, the rise of agentic AI, accelerating business speed, and heightened vendor M&A — are forcing changes in strategy. CISOs are simplifying tech stacks, increasing automation and outsourcing, and deploying AI for detection and response while wrestling with new authentication/authorization gaps. Vendor viability and consolidation now factor into resilience planning.

read more →

Mon, September 15, 2025

Amazon Connect Cases Adds Date Filters to List View

📅 Amazon Connect Cases now supports date range filters in the case list view, enabling managers and agents to narrow case lists by creation, modification, or upcoming SLA windows. Users can filter cases created in the last 30 days, view items modified in the past 24 hours, or surface cases with potential SLA breaches in the next two days for proactive triage. Amazon Connect Cases is available in multiple AWS regions and documentation is provided to get started.

read more →

Mon, September 15, 2025

Check Point Named Leader in 2025 IDC MarketScape Report

🔒 Check Point has been named a Leader in the 2025 IDC MarketScape for Worldwide Enterprise Hybrid Firewalls, reflecting the maturity of its Hybrid Mesh Firewall capabilities. The recognition highlights the company's focus on unifying, automating and simplifying security operations at scale. IDC reports the firewall market reached $12.3 billion in 2024, underscoring the importance of resilient, centralized protection. Enterprises face wider attack surfaces and operational complexity that such platforms aim to mitigate.

read more →

Mon, September 15, 2025

Browser-Based Attacks: Six Threats Security Teams Must Know

🔒 Browser-targeted attacks are rising as adversaries treat the browser as the primary access point to cloud services and corporate data. The article defines browser-based attacks and enumerates six high-risk techniques: credential and session phishing, ClickFix-style copy-and-paste exploits, malicious OAuth consent flows, rogue extensions, malicious file delivery, and credential reuse where MFA gaps exist. These vectors are effective because modern work happens in decentralized SaaS environments and across many delivery channels, making traditional email- and network-centric defenses less reliable. The piece highlights visibility gaps for security teams and points to vendor platforms such as Push Security that claim to provide in-browser detection and remediation for AiTM phishing, OAuth abuse, and session hijacking.

read more →

Mon, September 15, 2025

OIG: CISA Wasted Millions and Mismanaged Incentives

🔍 The DHS Office of Inspector General (OIG) audit found that CISA misused federal funds and undermined its mission by broadly administering the Cyber Incentive program. The review identified 240 recipients in non-cyber support roles, poor record-keeping in OCHCO, and $1.4m in undocumented back pay among more than $138m disbursed since 2020. Payments typically ranged from $21,000 to $25,000 annually per person, more than 40% of staff received incentives, and the OIG issued eight recommendations to tighten eligibility, tracking, governance and recovery procedures; CISA has concurred with all recommendations.

read more →

Mon, September 15, 2025

Amazon Managed Service for Prometheus Adds 11 Regions

📢 Amazon Managed Service for Prometheus is now generally available in 11 additional AWS regions, including Asia Pacific (Jakarta, Hyderabad, Osaka, Melbourne, Taipei), Canada West (Calgary), Europe (Spain), Israel (Tel Aviv), Mexico (Central), Middle East (Bahrain), and US West (N. California). The fully managed, Prometheus-compatible monitoring service makes it easier to collect, store, query, and alarm on operational metrics at scale. Customers can send up to 1 billion active metrics to a single workspace and create multiple workspaces per account to partition workloads. See the AWS user guide or product documentation for the full list of supported regions and configuration details.

read more →

Mon, September 15, 2025

Microsoft: September Windows Updates Break SMBv1 Shares

⚠️Microsoft confirmed that the September 2025 Windows security updates can break connections to SMBv1 shares when NetBIOS over TCP/IP (NetBT) is used. The issue affects client releases (Windows 11 24H2/23H2/22H2, Windows 10 22H2/21H2) and server releases (Windows Server 2025, 2022) and may occur if either the SMB client or server has the update. As a temporary workaround, administrators are advised to allow SMB traffic on TCP port 445 so Windows can switch from NetBT to TCP. Microsoft is investigating and developing a fix.

read more →

Mon, September 15, 2025

Fraudulent Account Created in Google's LERS Portal

🔒 Google has confirmed that a fraudulent account was created in its Law Enforcement Request System (LERS) and has been disabled. The company says no requests were made and no data was accessed. The claim was posted by a group calling itself Scattered Lapsus$ Hunters, which also alleged access to the FBI's eCheck system; the FBI declined to comment. The group has a history of high-profile Salesforce-related thefts and has publicly taunted law enforcement and security researchers.

read more →

Mon, September 15, 2025

Marine Corps Cuts ATO Delays with DevOps and Agile

🚀 Operation StormBreaker transformed how Marine Corps Community Services (MCCS) develops and authorizes IT. By creating a Marine Corps–authorized landing zone in AWS and pairing it with the Department of the Navy’s RAISE platform, MCCS implemented CI/CD pipelines and automated security checks to push security left. The result: ATOs that once took 18 months can now be granted in a day, saving roughly $1M per system and improving digital services for Marines and families.

read more →

Mon, September 15, 2025

Darknet Drug Shipping Ring Dismantled on German–Dutch Border

🚓 Law enforcement dismantled a darknet drug shipping operation across the German–Dutch border following an extensive IT-led investigation. Three men, aged 33, 39 and 40, are suspected of selling ecstasy and cocaine on darknet marketplaces and using border-area mailboxes to forward shipments into Germany. Searches of three residences and a boxing studio yielded multi-million-euro quantities of drugs, a firearm, five-figure cash, and numerous electronic devices that will now undergo forensic analysis.

read more →

Mon, September 15, 2025

Your SOC as the Parachute: Engineering for Resilience

🪂The SOC is framed as the parachute organisations rely on when breaches occur. Too many SOCs are under‑specified and reactive—drowned in alerts and tools that add complexity rather than resilience. The author calls for Swiss engineering: over‑specified, tested processes, rehearsed responses, and anticipatory defence grounded in threat modelling and behavioural context. Vendors and AI can assist, but organisations must own priorities, rehearse decision making, and build muscle memory.

read more →

Mon, September 15, 2025

Nine Essential Open-Source Security Tools for Teams

🔒 This article highlights nine widely used open-source security tools that help defenders identify vulnerabilities, analyze network traffic, perform forensic investigations, and manage threat intelligence. It stresses community-driven development and transparency as core advantages of open-source solutions and notes that independent review often speeds discovery and remediation. Representative tools covered include ZAP, Wireshark, BloodHound, Autopsy, MISP, Let's Encrypt, GnuPG, Yara and osquery, with attention to extensibility, multi-platform support, and practical deployment considerations for security teams.

read more →

Mon, September 15, 2025

Stop Waiting on NVD: Get Real-Time Vulnerability Alerts

🛡️SecAlerts provides real-time vulnerability alerts that avoid the publication delays commonly associated with NVD by aggregating signals from 100+ sources including vendors, researchers, forums and blogs. The service uses three core components — Stacks (software inventories and SBOMs), Channels (Email, Slack, Teams, Webhook) and Alerts (custom filters for Severity, Known Exploited, EPSS, Trending) — to deliver only relevant notifications. A central Dashboard surfaces affected software, extended metadata and reference links, while Properties enable multi-tenant views useful for MSPs and departments.

read more →

Mon, September 15, 2025

Ten Career Pitfalls That Can Derail Today's CISOs Now

🔒 CISOs face many behavioral and strategic traps that can stall or end careers if not addressed. Leaders, coaches and consultants identify ten common mistakes — from failing to align security with business priorities and treating security as a pure technology function, to reflexively saying no, enforcing rigid rules, misunderstanding AI, lacking transparency, not networking, and mishandling incidents. The article emphasizes becoming an enabler, tying controls to ROI, communicating clearly, and rehearsing response plans to build resilience.

read more →