All news in category “Vendor and Hyperscaler Watch”

🛡️ Microsoft says AI-driven discovery is increasing the pace of vulnerability identification in Windows, leading to a likely rise in monthly security updates. The company uses its MDASH system to scan critical binaries and validate potential issues with multiple AI models, then runs a Windows-specific validation pipeline to reduce false positives. Microsoft also applies AI to accelerate triage, suggest fixes, and find similar bugs, while keeping humans in the review loop. The firm is updating its Secure Development Lifecycle to address AI-enabled attack techniques as adversaries also leverage AI.

🚀 Microsoft announces general availability updates to Microsoft Foundry, combining frontier models, a production agent runtime, enterprise identity and security controls, and Microsoft 365 distribution into one platform. The release includes GPT-5.6 models across global regions and the new Asia-Pacific Data Zone for regional data processing. Developers can build in GitHub Copilot or VS Code and deploy agents using Foundry toolkits and SDKs.

🔒 GitHub released npm v12 which disables install scripts by default and deprecates 2FA-bypass granular access tokens. The update makes lifecycle scripts, Git dependencies, and remote URL deps opt-in, requiring an explicit approval workflow and an allowlist committed to package.json. It also restricts GAT capabilities for account and publishing actions, with staged publishing and OIDC recommended for automation.

🛡️ Cloud Run sandboxes are now in public preview, offering a native, secure, and ultra-fast runtime to execute untrusted code and agent workloads in milliseconds. These lightweight, isolated execution boundaries can spawn within existing Cloud Run service instances and enforce credential isolation, deny-by-default network egress, and a safe read-only filesystem overlay. Enabling sandboxes requires a single deployment flag and integrates with the Agent Development Kit and ComputeSDK for streamlined use.

🤖 AWS Partner Central introduces Partner Lead Prospecting, an AI-powered capability for ACE-eligible AWS Partners to accelerate lead conversion. Building on earlier lead enrichment features, it generates personalized sales plays, call scripts, and email outreach tailored to the customer's industry and each partner's solutions. The feature integrates a partner's portfolio into prospecting content to reduce manual research and drafting. Partner Lead Prospecting is available globally to all ACE-eligible partners via the Leads page or the MCP server.

🛠️ Amazon SageMaker Unified Studio now supports custom asset types for IAM-based domains, allowing administrators to catalog diverse asset formats such as medical images, PowerBI dashboards, or PDF reports. Administrators define a type with name, description, and optional metadata forms; assets created from those types can include glossary terms and README documentation. Published assets are discoverable and subscribable through the same governed workflow used for other catalog items and are available in all Regions where Unified Studio is offered.

🔍 Amazon SageMaker HyperPod now supports deep health checks for Slurm clusters created with continuous provisioning, enabling proactive verification of GPU accelerator health on running instances. Continuous provisioning allows asynchronous scaling of instance groups without all-or-nothing failures, and deep health checks validate hardware as nodes come online. Results and progress are visible via the SageMaker console and APIs, and failing instances are isolated and recovered automatically.

🚀 Amazon SageMaker Feature Store now supports high-throughput ingestion and record discovery with new APIs and offline-store naming options. Data scientists can use BatchWriteRecord to write multiple records across feature groups in one request, and ListRecords to page through record identifiers without prior knowledge. Offline-store configuration now allows creating Glue and Iceberg tables with custom names, simplifying cataloging and management.

🔁 Amazon MSK Replicator now supports replication from external Apache Kafka clusters to Amazon MSK Standard brokers. This expands prior capability that targeted MSK Express brokers and enables migration, disaster recovery, and hybrid or multi-cloud data distribution. The feature supports SASL/SCRAM and mutual TLS (mTLS) for authenticating to external clusters and is available in all Regions where MSK Replicator is offered. It preserves topic names, prevents replication loops, and syncs consumer group offsets bidirectionally.

🔒 The UK’s NCSC and DSIT unveiled a blueprint called Cyber Shield to deploy autonomous AI agents that detect and neutralize cyberattacks at machine speed. The plan uses cooperating “red” and “blue” agents to identify weaknesses, detect threats and progressively automate remediation while operating under organizational control. The initiative emphasizes explainable and federated AI, industry partnerships, and a staged rollout beginning with government and critical sectors.

🛡️ AWS Config has expanded its set of managed rules with 191 additional checks covering services such as Amazon Bedrock, SageMaker, ECS, EKS, RDS, Redshift, S3, and CloudTrail. The new rules evaluate encryption, logging, public access, network security, data protection, and operational best practices. You can deploy rules individually or as part of a conformance pack in supported AWS Regions.

📰 Microsoft will remove the OWA Light experience from on-premises Exchange Server in an upcoming update. The Exchange Team says retiring OWA Light reduces legacy surface area, simplifies engineering, and lets them focus on the full Outlook on the web experience. Administrators can proactively disable OWA Light via PowerShell using Set-OwaMailboxPolicy and Set-OwaVirtualDirectory commands. The change is expected in August 2026 after OWA Light was deprecated in August 2024.

🛠️ This blog explains how to configure GKE Autopilot clusters to use GKE managed DRANET for GPU and TPU workloads. It outlines the setup flow: create a VPC, deploy an Autopilot cluster, define a custom ComputeClass, create a ResourceClaimTemplate for RDMA (GPUs) or netdev (TPUs), and deploy workloads that reference those resources. Examples and YAML snippets demonstrate GPU and TPU ComputeClasses, resource claim templates, and a deployment that binds pods to accelerators.

🧩 AWS Builder Center now offers free, time-limited sandbox environments that builders can request directly from eligible workshops. These sandboxes remove the need for a personal AWS account, credit card, or worry about unexpected charges, allowing safe deployment of resources and experimentation in a pre-provisioned AWS account. Each sandbox grants 8 hours of access from activation, auto-cleans afterward, and most are ready within 15 minutes. Builders may request one sandbox per week (resets Sunday), and availability will expand to more workshops over time.

🔔 Amazon SageMaker Unified Studio Workflows now includes 19 new operators for Amazon Bedrock, Amazon S3 Tables, S3 Vectors, AWS Glue Data Catalog, and Amazon MWAA Serverless. These operators let users add tasks via the visual workflow creator to orchestrate services without writing custom integration code. The capabilities include managing Bedrock guardrails, provisioning and deleting S3 Tables and S3 Vectors, managing Glue Data Catalog assets, and triggering MWAA Serverless runs. This feature is available in all Regions where SageMaker Unified Studio is offered.

🔍 AWS Security Hub now includes Network Scanning to identify resources that are actually reachable from the public internet. The feature probes public IPs, VMs, and load balancers across AWS and Azure, detects reachable ports, and identifies services running behind them. Findings are created per reachable port and correlated by Security Hub Exposures to assess broader risk. Existing customers can enable the feature per account, region, or organization; it is enabled by default for new customers and included with Security Hub Essentials at no extra cost in supported commercial Regions.

🚀 C4N is Google Cloud’s new network- and block-storage-optimized Compute Engine instance family, now generally available after its Next ’26 preview. Built on a custom Titanium offload architecture and 5th Gen Intel Xeon CPUs, C4N delivers up to 400 Gbps network bandwidth, 95M PPS, and up to 25 GiB/s with Hyperdisk Extreme. It targets network-intensive, storage-heavy, and latency-sensitive workloads without requiring premium add-ons.

🚀 Amazon EC2 M8gd and R8gd instances are now available in additional AWS Regions, including Europe (Paris, Stockholm, Milan) and Asia Pacific (Mumbai, Osaka, Melbourne, Singapore, Taipei). These instances use AWS Graviton4 processors, offer up to 11.4 TB of local NVMe SSD storage, and deliver up to 30% better performance versus Graviton3. They support up to 50 Gbps network and 40 Gbps Amazon EBS bandwidth, with configurable bandwidth weighting and select sizes offering Elastic Fabric Adapter support.

🚀 Amazon Redshift now supports Graviton-based RG instances on the trailing track (P201) as of July 7, 2026. Customers can choose rg.4xlarge and rg.xlarge instance types for production workloads prioritizing stability, using the AWS Management Console, AWS CLI, or AWS SDKs. RG instances deliver up to 2.4x faster query performance than RA3 at 30% lower price per vCPU. Provision a new cluster or resize an existing cluster to migrate to RG on the trailing track.

🔔 Amazon Aurora DSQL change data capture (CDC) is now generally available, enabling real-time streaming of database changes to Amazon Kinesis Data Streams for event-driven architectures and data integration workflows. Aurora DSQL CDC captures insert, update, and delete operations as change events and delivers them to Kinesis Data Streams with no infrastructure to manage. Use CDC to synchronize microservices, trigger AWS Lambda, or route changes to Amazon S3, Amazon Redshift, and Amazon OpenSearch Service via Amazon Data Firehose. CDC streaming is available in all Regions where Aurora DSQL is offered and is designed to have zero impact on database workload performance.