Autonomous Defense, Cloudflare Outage, and a Chrome 0‑Day

Cloudflare outage (18 Nov 2025): feature file duplication

⚠️ On 18 November 2025, Cloudflare experienced a major outage after a permissions change in a ClickHouse database caused duplicated metadata to be emitted into a Bot Management feature file, doubling its size. The oversized file exceeded a preallocated feature limit in the core proxy, triggering a Rust panic and widespread HTTP 5xx errors. Cloudflare halted propagation, restored a known-good file, and restarted the proxy; services were largely restored by 14:30 UTC and fully recovered by 17:06 UTC. The company apologized and pledged architectural and process hardening to prevent recurrence.

AWS launches EC2 P6-B300 with NVIDIA Blackwell Ultra

🚀 Amazon Web Services has announced general availability of Amazon EC2 P6-B300 instances powered by NVIDIA Blackwell Ultra B300 GPUs. The p6-b300.48xlarge delivers eight GPUs, 2.1 TB of high-bandwidth GPU memory, 6.4 Tbps EFA networking, 300 Gbps ENA throughput, and 4 TB of system memory. It targets training and deploying trillion-parameter foundation models and LLMs, offering higher memory, compute, and networking versus P6-B200.

Azure Introduces Copilot Agents and AI Infrastructure

🚀 At Microsoft Ignite 2025, Microsoft unveiled a suite of Azure infrastructure and AI operational innovations built for scale, reliability, and security. Azure Copilot introduces an agentic operations model with six specialized agents—migration, deployment, optimization, observability, resiliency, and troubleshooting—designed to automate routine cloud management while enforcing RBAC and policy. The release also highlights new AI datacenter architecture (Fairwater), deployment of NVIDIA GB300 GPUs at scale, and platform improvements like Azure Boost and AKS Automatic to accelerate performance and reduce operational overhead.

Anthropic Claude Models Available in Microsoft Foundry

🚀 Microsoft announced integration of Anthropic's Claude models into Microsoft Foundry, making Azure the only cloud to provide both Claude and GPT frontier models on a single platform. The release brings Claude Haiku 4.5, Sonnet 4.5, and Opus 4.1 to Foundry with enterprise governance, observability, and deployment controls. Foundry Agent Service, the Model Context Protocol, skills-based modularity, and a model router are highlighted as tools to operationalize agentic workflows for coding, research, cybersecurity, and business automation. Token-based pricing tiers for the Claude models are published for standard deployments.

Windows 11 Introduces Cloud Rebuild and PITR Recovery

☁️ Microsoft unveiled two Windows 11 recovery capabilities — Point-in-Time Restore (PITR) and Cloud Rebuild — designed to reduce downtime and simplify device recovery. PITR builds on System Restore by capturing full system snapshots, enabling admins and users to roll a device back to a known-good state, including local files and applications. Cloud Rebuild allows remote reinstallation through Intune, leveraging Autopilot, OneDrive, and Windows Backup for Organizations to restore settings and data.

Google patches V8 zero-day in Chrome; admins urged

⚠️ Google released an emergency patch for a high‑severity Type Confusion vulnerability in the V8 JavaScript engine (CVE-2025-13223), which the company says is being exploited in the wild. The flaw, rated CVSS 8.8 and discovered by Clément Lecigne of Google TAG, affects Chromium‑based browsers and can enable heap corruption and potential code execution. Administrators should prioritize updating Chrome to the patched 142.0.7444.175/.176 builds. A second V8 issue, CVE-2025-13224, is also fixed.

TimesFM Integration Brings Forecasting to BigQuery

🕒 Google is integrating the TimesFM time-series foundation model into BigQuery and AlloyDB, enabling zero-shot forecasting on customer data without retraining. AI.FORECAST and AI.EVALUATE are now Generally Available in BigQuery, while AI.DETECT_ANOMALIES is in public preview. TimesFM 2.5 offers improved accuracy and lower latency, supports dynamic context windows up to 15K, and can return historical data with forecasts. AlloyDB preview lets users call TimesFM endpoints hosted on Vertex AI so operational data can be forecasted in-place, preserving data residency and reducing export overhead.

Azure Ignite 2025: Azure's Agentic AI and Data Innovations

🚀 At Microsoft Ignite 2025, Azure introduced a coordinated set of agentic and data-first capabilities to accelerate enterprise AI adoption. Announcements include Microsoft Agent Factory (available), previews of Fabric IQ and Foundry IQ, expanded Foundry model choices (Anthropic, Cohere), and new database offerings like Azure HorizonDB and Azure DocumentDB. The updates emphasize unified data, model choice, and integrated security to simplify building, running, and governing AI agents at scale.

Microsoft Databases and Fabric: Unified AI Data Estate

🧠 Microsoft details a broad expansion of its database portfolio and deeper integration with Microsoft Fabric to simplify data architectures and accelerate AI. Key launches include general availability of SQL Server 2025, GA of Azure DocumentDB (MongoDB-compatible), the preview of Azure HorizonDB, and Fabric-hosted SaaS databases for SQL and Cosmos DB. OneLake mirroring, Fabric IQ semantic modeling, expanded agent capabilities, and partner integrations (SAP, Salesforce, Databricks, Snowflake, dbt) are positioned to deliver zero-ETL analytics and operational AI at scale.

Ambient and Autonomous Security for the Agentic Era

🛡️ At Microsoft Ignite 2025, Microsoft set out an ambient, autonomous security approach for the emerging agentic era and announced a suite of tools to observe, secure, and govern AI agents and apps. The centerpiece is Microsoft Agent 365, a control plane providing an Entra-based registry, access controls, visualization, and integrations with Defender, Entra, and Purview to detect prompt-injection, prevent leakage, and enable auditing. Microsoft also expanded platform protections, enhanced Copilot data controls in Purview, and positioned Microsoft Sentinel and Security Copilot as agentic security pillars for detection and response.

Gemini 3 Brings Multimodal and Agentic AI to Enterprise

🤖 Google has made Gemini 3 available to enterprises and developers via Gemini Enterprise and Vertex AI, bringing advanced multimodal reasoning and agentic capabilities to production teams. The model can analyze text, images, video, audio, and code together, supports a 1M-token context window, and improves frontend generation, legacy code migration, and long-running tool orchestration. Early partners report faster diagnostics, richer UI prototypes, and more reliable automation across business workflows.

Cloudflare outage disrupts global network services

⚠️ Cloudflare is investigating an outage that has produced widespread 500 internal server errors and impacted its Dashboard and API, disrupting access to numerous customer websites and platforms. The company first reported support portal availability issues and then an incident at 11:48 UTC affecting the Cloudflare Global Network, with multiple European nodes observed offline. Downdetector logged tens of thousands of reports, and Cloudflare says it is working to mitigate the incident; partial recovery has been reported for Access and WARP while remediation continues for application services.

AWS offers flat-rate CloudFront plans with built-in security

🔒 AWS is introducing flat-rate pricing plans for CloudFront that bundle global CDN delivery with built-in security (WAF, DDoS protection), Route 53 DNS, CloudWatch Logs ingestion, serverless edge compute, and monthly S3 storage credits. Plans eliminate overage charges so traffic spikes or attacks won’t trigger surprise fees. Tiers include Free, Pro ($15), Business ($200) and Premium ($1,000), and pay-as-you-go remains an option.

Microsoft Foundry: Modular, Interoperable Secure Agent Stack

🔧 Microsoft today expanded Foundry, its platform for building production AI apps and agents, with new models, developer tools, and governance controls. Key updates include broader model access (Anthropic, Cohere, NVIDIA), a generally available model router, and public previews for Foundry IQ, Agent Service features (hosted agents, memory, multi-agent workflows), and the Foundry Control Plane. Foundry Tools and Foundry Local bring real-time connectors and edge inference, while Managed Instance on Azure App Service eases .NET cloud migrations.

Prisma AIRS Integration with Azure AI Foundry for Security

🔒 Palo Alto Networks announced that Prisma AIRS now integrates natively with Azure AI Foundry, enabling direct prompt and response scanning through the Prisma AIRS AI Runtime Security API. The integration provides real-time, model-agnostic threat detection for prompt injection, sensitive data leakage, malicious code and URLs, and toxic outputs, and supports custom topic filters. By embedding security into AI development workflows, teams gain production-grade protections without slowing innovation; the feature is available now via an early access program.

Silent FortiWeb Patch Raises Alarm as Critical Flaw Exploited

🔒 Fortinet's FortiWeb appliances are affected by a critical vulnerability tracked as CVE-2025-64446 that researchers say was exploited in the wild before an official advisory. The issue chains a relative path traversal to an internal CGI backend with an HTTP_CGIINFO header authentication bypass that allows unauthenticated admin impersonation and potential remote code execution. Fortinet released fixes in multiple 7.x and 8.x maintenance updates and recommends disabling HTTP/HTTPS on internet-facing management interfaces if upgrades cannot be applied immediately.

Amazon Bedrock adds Priority and Flex inference tiers

🔔 Amazon Bedrock introduces two new inference tiers—Priority and Flex—to help customers balance cost and latency for varied AI workloads. Flex targets non-time-critical jobs like model evaluations and summarization with discounted pricing and lower scheduling priority. Priority offers premium performance and preferential processing (up to 25% better OTPS vs. Standard) for mission-critical, real-time applications. The existing Standard tier remains available for general-purpose use.

OpenSearch Serverless: CloudTrail data-plane audit logging

🔒 Amazon has added detailed audit logging for OpenSearch Serverless data-plane requests through AWS CloudTrail. Customers can now record and retain user actions on collections — including authorization attempts, index changes, and search queries — to support compliance and incident investigations. Logs can be filtered with read-only or write-only options or captured using advanced event selectors for granular control. Data events are delivered to Amazon S3 and can be forwarded to Amazon CloudWatch Events for real-time monitoring and response.

Microsoft to Natively Integrate Sysmon in Windows 11

🛡️ Microsoft will integrate Sysmon natively into Windows 11 and Windows Server 2025, removing the need to deploy the standalone Sysinternals tool. The built-in functionality will preserve Sysmon’s capabilities, including support for custom configuration files and advanced event filtering, and logs events to the Windows Event Log. Administrators can enable it via Optional Features or run sysmon -i (or sysmon -i <config>) to load a custom configuration, and updates will be delivered through Windows Update to simplify management and improve coverage in large environments.

AWS Releases Responsible AI and Updated ML Lenses at Scale

🔔 AWS has published one new Responsible AI lens and updated Generative AI and Machine Learning lenses to guide safe, secure, and production-ready AI workloads. The guidance addresses fairness, reliability, and operational readiness while helping teams move from experimentation to production. Updates include recommendations for Amazon SageMaker HyperPod, Agentic AI, and integrations with Amazon SageMaker Unified Studio, Amazon Q, and Amazon Bedrock. The lenses are aimed at business leaders, ML engineers, data scientists, and risk and compliance professionals.

Security Copilot Agents Included with Microsoft 365 E5

🛡️ Microsoft is including Security Copilot agents in Microsoft 365 E5, embedding AI-driven assistants across Defender, Entra, Intune, and Purview to accelerate investigations and automate routine tasks. The rollout begins today for existing Security Copilot customers on E5 and will expand to all E5 tenants in the coming months with a 30-day notification. The announcement adds 12 Microsoft-built preview agents, 30+ partner agents, and support for customer-built agents to tailor workflows.

AWS Transfer Family Terraform Module Adds Malware Scanning

🛡️ AWS has updated the Transfer Family Terraform module to support automated malware scanning workflows for files transferred to S3. The module provisions GuardDuty S3 Protection–based scan pipelines, dynamic routing based on results, and threat notifications in a single deployment. It preserves folder structure, allows granular S3 prefix targeting, and helps ensure only verified clean files reach applications and data lakes.

AWS EC2 I7ie Instances Arrive in Singapore Region Now

🚀 Amazon Web Services (AWS) has launched EC2 I7ie instances in the Asia Pacific (Singapore) Region. Designed for large storage I/O–intensive workloads, I7ie pairs 5th Gen Intel Xeon processors with 3rd‑generation AWS Nitro SSDs to deliver up to 40% better compute and up to 65% improved real‑time storage performance versus I3en. Instances support up to 120 TB local NVMe density, up to twice the vCPU and memory of prior generations, nine size options, and up to 100 Gbps networking with 60 Gbps EBS bandwidth.

Amazon Polly adds five voices and three Asia Pacific regions

🎧 Amazon Polly now offers five new Generative TTS voices—Austrian German (Hannah), Irish English (Niamh), Brazilian Portuguese (Camila), Belgian Dutch (Lisa), and Korean (Seoyeon)—bringing the Generative engine to thirty-one voices across twenty locales. The Generative engine is generally available in three new Asia Pacific regions: Asia Pacific (Seoul), Asia Pacific (Singapore), and Asia Pacific (Tokyo), and all Generative voices are now available in US East (N. Virginia), Europe (Frankfurt), and US West (Oregon). These updates expand Amazon Polly's managed text-to-speech capabilities for conversational AI and speech content creation.

Google fixes new Chrome zero-day exploited in attacks

🔒 Google released an emergency update to address a newly discovered Chrome zero-day, CVE-2025-13223, which is being actively exploited. The high-severity flaw stems from a type confusion vulnerability in Chrome's V8 JavaScript engine and was reported by Clement Lecigne of Google's Threat Analysis Group. Patches are available in versions 142.0.7444.175/.176 for Windows, 142.0.7444.176 for macOS, and 142.0.7444.175 for Linux; users should check About Google Chrome and relaunch to apply the update.

Active Threat Defense Now Default in AWS Network Firewall

🔒 Starting today, AWS Network Firewall enables active threat defense by default in alert mode when you create new firewall policies in the AWS Management Console. The feature provides automated, intelligence-driven visibility into ongoing threat activities and the indicator groups, types, and names you are protected against. You can switch to block mode to automatically prevent suspicious traffic or disable the feature entirely. AWS verifies threat indicators to reduce false positives and the capability is available in all Regions, including AWS GovCloud (US) and China.

AWS Lambda Adds Python 3.14 Managed Runtime Support

🔔 AWS Lambda now supports Python 3.14 for both managed runtimes and as a container base image. AWS will automatically apply updates to the managed runtime and base image as they become available, reducing maintenance overhead. The runtime is available in all Regions, including AWS GovCloud (US) and China Regions, and is supported for Lambda@Edge in applicable Regions. Developers can deploy using the Lambda console, AWS CLI, AWS SAM, AWS CDK, and CloudFormation, and Powertools for AWS Lambda (Python) also supports Python 3.14.

CloudFormation introduces drift-aware change sets for IaC

🔁 AWS CloudFormation now offers drift-aware change sets to detect and reconcile configuration drift by comparing a new template, the last-deployed template, and the live infrastructure state. The feature lets you preview deployment impacts on drifted resources, avoid unintended overwrites, and revert out-of-band changes. Create a change set in the console as “Drift-aware” or call CreateChangeSet with --deployment-mode REVERT_DRIFT. Available in Regions where CloudFormation is offered.

AWS launches Storage-Optimized EC2 I7i instances in regions

🚀 AWS has made high performance, storage-optimized Amazon EC2 I7i instances available in Asia Pacific (Melbourne, Mumbai, Osaka) and Middle East (UAE) regions. Powered by 5th‑generation Intel Xeon Scalable processors and 3rd‑generation AWS Nitro SSDs, I7i delivers up to 23% better compute performance and more than 10% better price performance versus prior I4i instances. I7i offers up to 45 TB of NVMe storage with significant reductions in storage I/O latency and variability, supports torn-write prevention up to 16 KB blocks, and is offered in eleven sizes including bare metal, with up to 100 Gbps network and 60 Gbps EBS bandwidth.

Google Chrome fixes actively exploited V8 type bug

🛡️ Google has released emergency Chrome updates addressing two V8 engine type confusion flaws, including an actively exploited vulnerability tracked as CVE-2025-13223 (CVSS 8.8) that can lead to arbitrary code execution or crashes. The patch also fixes CVE-2025-13224 flagged by Google's AI agent Big Sleep and completes a set of seven zero-days addressed this year. Users should update Chrome to 142.0.7444.175/.176 (Windows/macOS/Linux) and apply fixes for other Chromium-based browsers when available.

Fortinet warns of FortiWeb zero-day being exploited

🚨 Fortinet has released security updates to remediate a new FortiWeb zero-day tracked as CVE-2025-58034, which the vendor says is being actively exploited in the wild. The vulnerability is an authenticated OS command injection (CWE-78) that can allow an attacker to execute code via crafted HTTP requests or CLI commands without user interaction. Fortinet confirmed observed exploitation and published fixes; administrators should upgrade affected FortiWeb appliances to the patched releases as soon as possible.

Microsoft Mitigates 15.72 Tbps IoT-Driven DDoS Attack

🛡 Microsoft automatically detected and mitigated a massive DDoS attack that peaked at 15.72 Tbps and roughly 3.64 billion packets per second against a single Australian endpoint. The traffic was attributed to a TurboMirai-class IoT botnet called AISURU, sourced from hundreds of thousands of compromised routers, cameras, and DVRs and launched from over 500,000 source IPs across multiple regions. Attackers used high-rate UDP floods with minimal source spoofing and random source ports, factors Microsoft said helped simplify traceback and provider enforcement. The incident underscores rising DDoS baselines as broadband speeds increase and IoT devices become more capable.

ShadowRay 2.0 Converts Exposed Ray Clusters to Miners

⚠ A global campaign named ShadowRay 2.0 is exploiting an unpatched code-execution flaw (CVE-2023-48022) in Ray clusters to deploy a self-propagating cryptomining botnet. Researchers at Oligo attribute the activity to an actor tracked as IronErn440, which uses AI-generated payloads submitted to Ray’s unauthenticated Jobs API. The malware deploys XMRig to mine Monero, establishes persistence via cron and systemd, and opens reverse shells for interactive control. Operators also throttle CPU use and conceal miners with deceptive names to evade detection.

METZ CONNECT EWIO2 Firmware Critical Vulnerabilities

🔒 METZ CONNECT released firmware updates addressing multiple critical vulnerabilities in EWIO2 devices that allow unauthenticated remote attackers to bypass authentication, upload and execute arbitrary code, and read PHP source files. The flaws include an authentication bypass, PHP remote file inclusion, unrestricted file uploads, path traversal, and improper access control. METZ CONNECT firmware 2.2.0 remediates these issues; administrators should schedule and install the update and ensure devices are not exposed to the internet.

Azure Mitigates Record 15.72 Tbps DDoS from IoT Botnet

🛡️ Microsoft Azure said it blocked a record 15.72 Tbps DDoS attack tied to the Aisuru IoT botnet that surged to roughly 3.64 billion packets per second and targeted a single cloud endpoint in Australia. The attacker launched extremely high-rate UDP floods from over 500,000 source IPs with minimal spoofing and random source ports. Azure DDoS Protection automatically detected and mitigated the traffic without disrupting customer workloads, and Microsoft urged organizations to validate internet-facing protections ahead of peak periods, noting systemic IoT security gaps.

French Pajemploi Reports Data Breach Affecting 1.2M

🔒 French social security service Pajemploi disclosed a data breach detected on November 14 that may have exposed personal information for up to 1.2 million registered home-based childcare workers and parents. Potentially exfiltrated data includes full names, place of birth, postal addresses, social security numbers, names of banking institutions, Pajemploi numbers, and accreditation numbers. The agency says IBANs, email addresses, phone numbers, and passwords were not accessed. Pajemploi notified CNIL and ANSSI, will inform affected individuals, and URSSAF warned of increased phishing and social engineering risks.

CISA Adds Fortinet FortiWeb Command Injection CVE Advisory

⚠️ CISA has added CVE-2025-58034, a Fortinet FortiWeb OS command code injection vulnerability, to its Known Exploited Vulnerabilities (KEV) Catalog after evidence of active exploitation. The agency recommends a reduced remediation timeframe of one week due to recent and ongoing exploitation and points to BOD 23-02 for steps to limit exposure from internet-accessible management interfaces. Although BOD 22-01 applies to Federal Civilian Executive Branch agencies, CISA strongly urges all organizations to prioritize timely remediation and vulnerability management for KEV entries.

Shelly Pro 3EM Out-of-Bounds Read Causes Reboots and DoS

⚠️ A remote-accessible out-of-bounds read vulnerability (CVE-2025-12056) in Shelly Pro 3EM can be triggered by a specially crafted Modbus request to force the device to access illegal memory addresses and reboot. CISA assigns a CVSS v4 score of 8.3 and warns this may result in a denial-of-service condition. Shelly did not respond to coordination; users should contact the vendor, keep devices updated, minimize network exposure, and follow recommended ICS defensive practices.

Tycoon 2FA Phishing Kit Undermines Legacy MFA Protections

🔐 Tycoon 2FA is a turnkey phishing kit that automates real-time MFA relays, enabling attackers to capture credentials, session cookies, and live authentication flows for Microsoft 365 and Gmail. It requires no coding skill, includes layered evasion (obfuscation, compression, bot filtering and debugger checks), and proxies MFA prompts so victims unknowingly authenticate attackers. The result undermines SMS, TOTP and push methods and can enable full session takeover. The article urges migration to phishing-resistant FIDO2 hardware and domain-bound biometric authenticators.

Iranian-backed UNC1549 Deploys TWOSTROKE and DEEPROOT

🛡️ Mandiant has linked suspected Iranian espionage actors to a sustained campaign by UNC1549 that deployed backdoors such as TWOSTROKE and DEEPROOT against aerospace, aviation, and defense organizations in the Middle East. Operating from late 2023 through 2025, the group abused trusted third parties and VDI sessions to pivot into customer environments and leveraged highly targeted, role‑relevant phishing. Observed operations combined credential theft, lateral movement, custom tunnellers and credential‑stealing utilities to execute long‑term reconnaissance and data exfiltration.

Snort3 Adds Severity-Based Rule Grouping for Flexibility

🔔 Cisco Talos has introduced a new Severity rule group for Snort3 in Cisco Secure Firewall, grouping detection rules by CVSS-derived severity tiers (low, medium, high, critical). Administrators can set coverage by time range — from the last two years up to all historical vulnerabilities — to balance detection depth and performance. This makes it easier to align rules with patch cycles, compliance needs, and organizational risk priorities while reducing manual tuning.

Amazon MSK Replicator expands to Hyderabad and Malaysia

🔁 Amazon Web Services has expanded MSK Replicator to the Asia Pacific (Hyderabad) and Asia Pacific (Malaysia) Regions, enabling customers to replicate streaming data between Amazon MSK clusters. MSK Replicator automates asynchronous replication and preserves Kafka metadata such as topic configurations, ACLs, and consumer offsets, while automatically scaling underlying resources. The feature simplifies failover and regional resilience and is available from the MSK console or CLI.

Windows 10 KB5072653 OOB Update Fixes ESU Install Errors

🛠️ Microsoft released the out-of-band update KB5072653 to address installation failures affecting the November Extended Security Update for Windows 10. The preparation package resolves 0x800f0922 (CBS_E_INSTALLERS_FAILED) errors and requires devices to run Windows 10 22H2 with the October cumulative update KB5066791. KB5072653 will be offered automatically via Windows Update; after installing and restarting, administrators should rerun Windows Update to deploy the November ESU update (KB5068781). Microsoft will also publish updated Scan Cab metadata for WSUS/SCCM customers who rely on cab files for compliance checks.

Tycoon 2FA Kit Exposes Global Collapse of Legacy MFA

🔐 The Tycoon 2FA phishing kit is a turnkey, scalable Phishing-as-a-Service that automates real-time credential and MFA relay attacks against Microsoft 365 and Gmail. It provisions fake login pages and reverse proxies, intercepts usernames, passwords and session cookies, then proxies the MFA flow so victims unknowingly authenticate attackers. The kit includes obfuscation, compression, bot-filtering, CAPTCHA and debugger checks to evade detection and only reveals full behavior to human targets. Organizations are urged to adopt FIDO2-based, hardware-backed biometric and domain-bound authentication to prevent such relay attacks.

AI-Enhanced Tuoni Framework Targets US Real Estate Firm

🔍 Morphisec observed an AI-enhanced intrusion in October 2025 that targeted a major US real estate firm using the modular Tuoni C2 framework. The campaign began with a Microsoft Teams impersonation and a PowerShell one-liner that spawned a hidden process to retrieve a secondary script. That loader downloaded a BMP file and used least significant bit steganography to extract shellcode, executing it entirely in memory and reflectively loading TuoniAgent.dll. Researchers noted AI-generated code patterns and an encoded configuration pointing to two C2 servers; Morphisec's AMTD prevented execution.

Stadtwerke Detmold Hit by Hacker Attack, IT Shutdown

🔒 Stadtwerke Detmold has reported a widespread IT outage following an apparent hacker attack that prompted the operator to take all systems offline. Online services are unavailable and the company cannot be reached by phone or email. The utility says the supply of drinking water, electricity, gas and district heating remains assured, and customers can report technical problems via a hotline. Authorities are investigating the incident and, so far, no ransom demand has been reported.

Schneider Electric PowerChute Serial Shutdown Fixes

🔒 Schneider Electric has released updates for PowerChute Serial Shutdown to address multiple vulnerabilities that may be exploited locally on the network. The issues include path traversal (CWE-22, CVE-2025-11565), excessive authentication attempts (CWE-307, CVE-2025-11566), and incorrect default permissions (CWE-276, CVE-2025-11567) with CVSS scores up to 7.8. Schneider Electric published version 1.4 with fixes for Windows and Linux; administrators should upgrade and apply recommended permissions and network isolation measures.

Defeating BLOCKADE SPIDER: Stopping Cross-Domain Attacks

🔒 CrowdStrike describes how OverWatch detected and disrupted BLOCKADE SPIDER, a financially motivated eCrime group that has used cross-domain techniques since at least April 2024 to access unmanaged systems, dump credentials, and deploy Embargo ransomware. By correlating endpoint, identity, and cloud telemetry in Falcon Next-Gen SIEM and Falcon Identity Threat Protection, analysts traced a compromised VPN service account and observed MFA bypass and AD manipulation. The account underscores the value of unified visibility to stop lateral movement and protect critical assets.

DoorDash Confirms October 2025 Customer Data Breach

🔒 DoorDash has confirmed a data breach in October 2025 that exposed customers' names, phone numbers, physical addresses and email addresses. The company said an employee was targeted in a social engineering scam that allowed unauthorized access, but there is currently no indication the data has been misused. DoorDash stated that sensitive identifiers and payment information were not accessed and that it has engaged an external firm, notified law enforcement, rolled out security enhancements and issued additional staff training.

Using Private NAT for Overlapping Private IP Spaces

🔒 Google Cloud's Private NAT enables secure private-to-private translation to connect networks with overlapping or non-routable IPv4 ranges without running NAT appliances. As a managed Cloud NAT feature, it delivers high availability, automatic scalability, and centralized control for hybrid and multi‑VPC topologies. The post includes practical gcloud examples and Network Connectivity Center use cases to guide implementation.

npm Malware Campaign Redirects Visitors to Fake Crypto Sites

🛡️ Researchers from the Socket Threat Research Team uncovered a new npm malware campaign operated by threat actor dino_reborn, distributed across seven packages that executed immediately and fingerprinted visitors. The packages used Adspect proxying and cloaking to distinguish researchers from victims, delivering branded fake CAPTCHAs and dynamic redirects to malicious crypto sites. Anti-analysis measures disabled developer tools and user interactions to hinder inspection.

Fortinet and CSI Launch Global Cybercrime Bounty Program

🛡️ Fortinet and Crime Stoppers International (CSI) have launched the Cybercrime Bounty program, a global initiative enabling secure, anonymous reporting of cybercriminal activity. Validated reports will feed Fortinet’s threat intelligence to support law enforcement investigations and potential prosecutions. The program scales deterrence by combining community-sourced tips with expert analysis, building on decades of Fortinet collaboration with INTERPOL and other public-private partners.

Amazon Redshift JIT ANALYZE for Apache Iceberg tables

📈 Amazon Redshift now supports Just‑In‑Time (JIT) ANALYZE for Apache Iceberg tables, automatically collecting table‑ and column‑level statistics during query execution. The feature uses intelligent heuristics and lightweight sketch data structures to determine when runtime statistics will improve optimizer decisions and to build high‑quality statistics on the fly. JIT ANALYZE is generally available in all AWS regions with Redshift and requires no configuration changes to begin improving query plans and performance.

CloudFormation adds pre-deploy validation and operation IDs

🔎 AWS CloudFormation now validates change sets for three common template errors—invalid property syntax, resource name conflicts with existing account resources, and S3 bucket emptiness constraints on delete—before provisioning begins. If validation fails, the change set status is marked 'FAILED' and includes detailed failure messages and property paths to pinpoint issues. Runtime failures can still occur during execution, so CloudFormation also groups stack events by a unique operation ID that you can view in the Console or via the describe-operation API to filter and diagnose failures quickly. Together these capabilities are intended to shorten deployment cycles and cut troubleshooting time from minutes to seconds.

Half a Million FTSE 100 Credentials Discovered Online

🔒 Security researchers from Socura and Flare found around 460,000 compromised credentials tied to FTSE 100 domains across clear- and dark-web crime communities, including 28,000 entries from infostealer logs. The report notes many companies had thousands of leaks and that password hygiene remains poor, with 59% having at least one user using 'password'. It recommends MFA, passkeys, password managers, conditional access and proactive leak monitoring.

Schneider Electric: Risky Cryptography in EcoStruxure

🔒 This advisory describes a cryptographic weakness in Schneider Electric's EcoStruxure Machine SCADA Expert and Pro-face BLUE Open Studio that could allow credential recovery from project files. An attacker with read access to Edge project or offline cache files can brute-force weak hashes to recover app-native or Active Directory passwords (CVE-2025-9317); the flaw requires local/file access and is not remotely exploitable. Apply 2023.1 Patch 1 immediately or implement recommended mitigations such as strict ACLs, strong project master passwords, removing embedded passwords, and following ICS cybersecurity best practices.

AWS Network Firewall Log Analysis Using OpenSearch

📊 The post describes a new Amazon CloudWatch and Amazon OpenSearch Service dashboard that simplifies analysis of AWS Network Firewall logs by removing previous multi-step setup and streamlining integration. It explains prerequisites, creating an OpenSearch integration and dashboard, selecting log groups, sync intervals, and IAM roles. The overview covers widgets, filters, CSV export, common use cases, and cost considerations to improve visibility and troubleshooting.

EC2 Auto Scaling adds synchronous LaunchInstances API

🔔 Today, EC2 Auto Scaling launched a synchronous LaunchInstances API that gives customers precise control over where instances are provisioned and provides immediate feedback on capacity availability. The API supports overrides for any Availability Zone and/or subnet in an Auto Scaling group and includes optional asynchronous retries to help reach desired capacity. It is available now in US East (N. Virginia), US West (Oregon), Europe (Ireland), and Asia Pacific (Singapore) at no additional cost beyond standard EC2 and EBS usage. Use the AWS CLI or SDKs to get started.

AWS Backup releases low-cost warm storage for S3 backups

🚀 AWS Backup introduces a low-cost warm storage tier for Amazon S3 backups that can cut storage costs by up to 30%. After S3 backup data resides in a vault for 60 days (configurable to a longer period), you can automatically move it to the new tier while preserving the same performance and features, including ransomware protection, recovery, and auditing. Automatic tiering can be enabled at the account, vault, or bucket level and is available in all Regions where AWS Backup for S3 is offered; a one-time transition fee applies.

AWS launches Supplementary Packages for Amazon Linux

📦 AWS announced the general availability of Supplementary Packages for Amazon Linux (SPAL), a curated repository offering thousands of pre-built EPEL9-compatible packages for Amazon Linux 2023 (AL2023). SPAL reduces the need to compile software from source, accelerating deployments and lowering operational overhead for developers, system administrators, and DevOps teams. Packages are derived from community EPEL9 sources with AWS applying security patches as they become available upstream. SPAL is available across all AWS Commercial Regions, including GovCloud and China.

The AI Fix #77: Genome LLM, Ethics, Robots and Romance

🔬 In episode 77 of The AI Fix, Graham Cluley and Mark Stockley survey a week of unsettling and sometimes absurd AI stories. They discuss a bioRxiv preprint showing a genome-trained LLM generating novel bacteriophage sequences, debates over whether AI should be allowed to decide life-or-death outcomes, and a woman who legally ‘wed’ a ChatGPT persona she named "Klaus." The episode also covers a robot's public face-plant in Russia, MIT quietly retracting a flawed cybersecurity paper, and reflections on how early AI efforts were cobbled together.

Amazon RDS Adds R8gd and M8gd for Optimized Reads Now

🚀 Amazon RDS now supports R8gd and M8gd instances for Optimized Reads on Amazon Aurora PostgreSQL and RDS for PostgreSQL, MySQL, and MariaDB, offering improved price-performance. Optimized Reads uses local NVMe-based SSDs to store ephemeral data (temporary tables and evicted pages) to reduce network I/O and boost read latency and throughput. AWS reports up to 165% better throughput and up to 120% better price-performance on R8gd versus R6g for Aurora PostgreSQL. These instances are available in select US, Europe, and Asia Pacific regions and can be enabled via the Console, CLI, or SDK.

Energy Sector Targeted by Hackers: Risks, AI & Cooperation

🔒 The energy sector faces a high and growing cyber threat, with attackers targeting OT systems, grid sensors and IoT endpoints to create cascading societal impacts. Critical vulnerabilities — notably in Siemens products — and increasing IT‑OT coupling widen the attack surface. The article stresses the need for end-to-end visibility, AI-driven early warning and anomaly detection, and stronger international cooperation, including NIS 2-aligned practices and active CERT coordination to build resilience.

Google Cloud designated as DORA critical ICT provider

🔒 Google Cloud EMEA has been designated a critical ICT third-party provider under the EU DORA. The designation acknowledges the systemic importance of financial entities using Google Cloud services and establishes a direct oversight channel with a Lead Overseer from the ESAs. Google Cloud commits to transparency, customer support for compliance, and collaboration to strengthen digital operational resilience across Europe. They provide resources like a Register of Information Guide and an ICT Risk Management Customer Guide to support customers' compliance journeys.

Researchers Detail Tuoni C2's Role in Real-Estate Attack

🔒 Cybersecurity researchers disclosed an attempted intrusion against a major U.S. real-estate firm that leveraged the emerging Tuoni C2 and red-team framework. The campaign, observed in mid-October 2025, used Microsoft Teams impersonation and a PowerShell loader that fetched a BMP-steganographed payload from kupaoquan[.]com and executed shellcode in memory. That sequence spawned TuoniAgent.dll, which contacted a C2 server but ultimately failed to achieve its goals. The incident highlights the risk of freely available red-team tooling and AI-assisted code generation being abused by threat actors.

Shelly Pro 4PM DoS Vulnerability (CVE-2025-11243)

⚠ A vulnerability in Shelly Pro 4PM (CVE-2025-11243) can cause device reboots and denial-of-service conditions. Due to insufficient input bounds checking in the device's JSON parser, specially crafted RPC requests can trigger memory overallocation and force a reboot. Devices running firmware prior to v1.6 are affected; CISA notes the exploit is reachable from adjacent networks with low attack complexity. Operators should update to v1.6.0 or later and limit network exposure.

Generative AI Drives Rise in Deepfakes and Digital Forgeries

🔍 A new report from Entrust analyzing over one billion identity verifications between September 2024 and September 2025 warns that fraudsters increasingly use generative AI to produce hyper-realistic digital forgeries. Physical counterfeits still account for 47% of attempts, but digital forgeries now represent 35%, while deepfakes comprise 20% of biometric frauds. The report also highlights a 40% annual rise in injection attacks that feed fake images directly into verification systems.

Amazon Redshift Adds SUPER Type for Case-Insensitive DBs

🧩 Amazon Redshift now supports the SUPER data type in databases configured with case-insensitive collation, enabling analytics on semi-structured and nested data alongside structured SQL types. Using PartiQL, users can query JSON and mixed data without additional normalization. The COLLATE function lets you explicitly control case sensitivity for SUPER columns. This capability is available in all AWS Regions, including AWS GovCloud (US).

Rethinking Identity in the AI Era: Building Trust Fast

🔐 CISOs are grappling with an accelerating identity crisis as stolen credentials and compromised identities account for a large share of breaches. Experts warn that traditional, human-centric IAM models were not designed for agentic AI and the thousands of autonomous agents that can act and impersonate at machine speed. The SINET Identity Working Group advocates an AI Trust Fabric built on cryptographic, proofed identities, dynamic fine-grained authorization, just-in-time access, explicit delegation, and API-driven controls to reduce risks such as prompt injection, model theft, and data poisoning.

Google Cloud launches 30-day Cloud SQL free trial edition

🚀 Google Cloud today announced a 30-day free trial for Cloud SQL, offering preconfigured MySQL and PostgreSQL instances to evaluate enterprise-grade features without upfront commitment. The trial exposes Performance, High Availability, and Data Cache capabilities and mirrors production-oriented SKUs (Enterprise and Enterprise Plus). Users can one-click upgrade to paid instances at any time; after 30 days instances are suspended and kept stopped for an additional 90 days at no charge. The offer is available in all Google Cloud regions.

Amazon RDS for Oracle: October 2025 RU and Spatial Fixes

🔔 Amazon RDS for Oracle now supports the Oracle October 2025 Release Update (RU) for 19c and 21c, and the corresponding Spatial Patch Bundle for 19c. AWS recommends upgrading because the RU includes six new security patches for Oracle database products, and the Spatial Patch Bundle provides important fixes and performance improvements for Oracle Spatial and Graph. You can apply the RU from the Amazon RDS Management Console or programmatically via the AWS SDK or CLI, and enable Automatic Minor Version Upgrade to install updates during your maintenance window. To deploy the Spatial Patch Bundle, select the 'Spatial Patch Bundle Engine Versions' checkbox when creating new instances or upgrade existing instances to engine version '19.0.0.0.ru-2025-10.spb-1.r1'.

Amazon RDS Supports MariaDB 10.6.24, 10.11.15, 11.4.9

🔔 Amazon RDS for MariaDB now supports community minor versions 10.6.24, 10.11.15, and 11.4.9. Customers are recommended to upgrade to these latest minor releases to address known security vulnerabilities and gain bug fixes, performance improvements, and new community features. You can enable automatic minor version upgrades or use Amazon RDS Managed Blue/Green deployments to apply updates during scheduled maintenance windows. See the Amazon RDS User Guide for upgrade and deployment details.

Microsoft fixes Windows 10 ESU update installation error

🔧 Microsoft acknowledged that the November Patch Tuesday update KB5068781 for Windows 10 (builds 19044.6575 and 19045.6575) could fail to install on commercial devices activated via Windows subscription activation through the Microsoft 365 admin center, producing error 0x800f0922. On Nov. 17 Microsoft issued a preparation package, KB5072653, to resolve the problem and allow deployment of the November security update. Administrators should verify the latest servicing stack update, run the Windows Update Troubleshooter, and, if needed, install the .msu manually via wusa.exe.

Thunderbird Gains Native Microsoft Exchange Support

📧 Thunderbird 145 introduces built-in support for Microsoft Exchange email via the Exchange Web Services (EWS) protocol, eliminating the need for third-party add-ons in Exchange-hosted environments. The client auto-detects account settings and uses Microsoft’s OAuth2 for authorization to simplify migration from Outlook. Initial capabilities include full folder listings, message synchronization, message operations (view, send, reply, forward, move, copy, delete), attachment handling, subject/body search and quick filtering for Microsoft 365 domains with standard OAuth2 and for on-premise Exchange using basic password authentication. The Thunderbird team says additional features such as calendar syncing, address book support, Microsoft Graph integration and expanded authentication options (NTLM, tenant-specific OAuth2) are planned but not yet available.

Sneaky 2FA Kit Adds BitB Pop-ups That Mimic Address Bar

🔒 Push Security says the Sneaky 2FA Phishing-as-a-Service kit now leverages Browser-in-the-Browser (BitB) pop-ups to impersonate Microsoft login pages and conceal malicious URLs. Victims first pass a Cloudflare Turnstile bot check before a fake "Sign in with Microsoft" flow is loaded in an embedded BitB window that exfiltrates credentials and session data. The campaign pairs conditional loading, developer‑tool blocking, obfuscation, and rapid domain rotation; organizations should tighten conditional access and users should avoid unknown links and browser extensions.

Validating Chrome Extensions: Organizational Security

🔒 This article by Stan Kaminsky reviews Athanasios Giatsos’ Security Analyst Summit 2025 talk and explains why malicious browser extensions are a major blind spot for organizations. It outlines how extensions can access cookies, local storage, proxy settings, clipboard and screen capture, enabling session and account theft, espionage, ad fraud and crypto theft, and why Manifest V3 reduces but does not eliminate risk. Practical controls described include formal extension policies and allowlists, disabling developer mode, version pinning and testing of updates, EDR and SIEM-based monitoring, and the use of specialized vetting tools for deeper analysis.

CISA 2015 Short-Term Extension Provides Temporary Relief

🛡️ The US Cybersecurity Information Sharing Act (CISA 2015) received a three-month extension in a Senate continuing resolution, preserving liability protections for voluntary threat sharing through the Automated Indicator Sharing (AIS) program until January 30, 2026. Cyber professionals broadly welcomed the move but called it a "temporary patch" and urged a longer-term renewal. Industry sources reported the lapse since September reduced federal-to-private sharing, while a Binalyze survey highlighted operational strains, estimating an average cost of $114,000 per hour of delayed incident response.

AI and Voter Engagement: Transforming Political Campaigning

🗳️ This essay examines how AI could reshape political campaigning by enabling scaled forms of relational organizing and new channels for constituent dialogue. It contrasts the connective affordances of Facebook in 2008, which empowered person-to-person mobilization, with today’s platforms (TikTok, Reddit, YouTube) that favor broadcast or topical interaction. The authors show how AI assistants can draft highly personalized outreach and synthesize constituent feedback, survey global experiments from Japan’s Team Mirai to municipal pilots, and warn about deepfakes, artificial identities, and manipulation risks.

CISA Issues Six New Industrial Control Systems Advisories

🔔 CISA released six Industrial Control Systems (ICS) advisories detailing current security issues, vulnerabilities, and potential exploits affecting multiple vendors and products. The advisories cover Schneider Electric products (including EcoStruxure Machine SCADA Expert, Pro-face BLUE Open Studio, and PowerChute Serial Shutdown), Shelly Pro devices, and METZ CONNECT hardware. One advisory is an update (B) to a prior Schneider Electric notice. Users and administrators are encouraged to review the technical details and apply recommended mitigations promptly.

Checkout.com Apologizes After Breach, Donates Ransom

🔒 Checkout.com publicly disclosed a breach after the ShinyHunters group accessed data from a legacy third‑party cloud storage system used prior to 2020, and issued an apology taking responsibility for the error. The company said fewer than 25% of current merchants were affected, confirmed no payment card data was taken, and refused the ransom demand. Instead of paying, it donated the ransom amount to Carnegie Mellon University and the University of Oxford Security Center to support research into cybercrime.

How AI Is Reshaping Enterprise GRC and Risk Control

🔒 Organizations must update GRC programs to address the rising use and risks of generative and agentic AI, balancing innovation with compliance and security. Recent data — including Check Point's AI Security Report 2025 — indicate roughly one in 80 corporate requests to generative AI services carries a high risk of sensitive data loss. Security leaders are advised to treat AI as a distinct risk category, adapt frameworks like NIST AI RMF and ISO/IEC 42001, and implement pragmatic controls such as traffic-light tool classification and risk-based inventories so teams can prioritize highest-impact risks without stifling progress.

Microsoft Teams adds false-positive threat report option

🔔 Microsoft is adding a user-driven false-positive reporting capability to Microsoft Teams, allowing users to flag chat or channel messages they believe were incorrectly marked as security threats. The feature began a targeted rollout in September and is scheduled to reach general availability worldwide by the end of November 2025. It will be available to organizations using Microsoft Defender for Office 365 Plan 2 or Microsoft Defender XDR, and accessible on desktop, mobile, and web. Administrators can enable or disable the feature from the Teams admin center or the Microsoft Defender portal.

Fine-tuning MedGemma for Breast Tumor Classification

🧬 This guide demonstrates step-by-step fine-tuning of MedGemma (a Gemma 3 variant) to classify breast histopathology images using the public BreakHis dataset and a notebook-based workflow. It highlights practical choices—using an NVIDIA A100 40 GB, switching from FP16 to BF16 to avoid numerical overflows, and employing LoRA adapters for efficient training. The tutorial reports dramatic accuracy gains after merging LoRA adapters and points readers to runnable notebooks for reproducibility.

AWS Builder Center Adds Searchable Workshops Catalog

📚 AWS Builder Center now hosts the Workshops Catalog, offering step-by-step, expert-authored tutorials that guide users through deploying and using AWS services in their own accounts. The catalog includes hundreds of workshops across categories like Machine Learning, Security, and Serverless, with filters for service, complexity (100–400), and duration. Content is localized to the Builder Center language preference and accessible with a free Builder ID for hands-on learning.

Automating Session Manager Preferences with CloudFormation

🔐 This post explains how to centrally manage AWS Systems Manager Session Manager preferences across multiple accounts and Regions using CloudFormation StackSets and an AWS Lambda function. The solution automates updates to the SSM-SessionManagerRunShell document, provisions optional logging destinations (Amazon S3 or CloudWatch Logs), and can create KMS keys for session and log encryption. It aims to reduce manual configuration errors and ensure consistent security and compliance at scale.

Amazon FSx for Lustre: Up to 5× Faster Directory Listing

🚀 Amazon FSx for Lustre now delivers up to 5× faster directory listing performance, reducing the time it takes to run ls and inspect file system contents. The improvement benefits interactive use cases such as home directories and source code repositories commonly used by ML researchers, data scientists, and developers. These gains are provided via the Lustre 2.15 client in all FSx regions; upgrade to 2.15 and apply the recommended client tunings to realize the speedups.

Meta Expands WhatsApp Security Research Effort

🛡️ Meta has provided selected long‑time bug bounty researchers with a new tool, WhatsApp Research Proxy, to streamline analysis of WhatsApp's network protocol and reduce barriers to in‑depth research. The company is also running a pilot that invites research teams to focus on platform abuse with internal engineering and tooling support. Meta said it has paid more than $25 million to over 1,400 researchers in 15 years and recently added anti‑scraping protections after a study showed an account‑enumeration technique able to map billions of users.