All news in category “Vendor and Hyperscaler Watch”

🚀 Today at Google Cloud Next, Google announced a more proactive Gemini Cloud Assist, an agentic cloud operations platform that embeds Gemini intelligence and enterprise context into the operational layer. It automates design-to-deployment workflows via a redesigned Application Design Center, supports infrastructure automation with gcloud, kubectl, and Terraform, and runs proactive multi-turn agents for troubleshooting and FinOps cost anomaly detection. The service also publishes its capabilities as MCP servers so teams can access design, operation, troubleshooting and optimization features directly from IDEs, CLIs, and third-party toolchains.

🚀 Google Cloud today launched Gemini Enterprise Agent Platform, the successor to Vertex AI designed to build, scale, govern, and optimize production-grade AI agents. The platform centralizes access to 200+ models via Model Garden, and provides visual and code-first tooling through Agent Studio and the Agent Development Kit (ADK). It adds a long-running Agent Runtime with Memory Bank, identity and registry services, and integrated security, simulation, and observability to accelerate and govern agent-driven workflows.

🚀 Google Cloud introduces its Agentic Data Cloud to remove the barrier between analytical history and live operational data, enabling real-time AI decisioning. By integrating AlloyDB, BigQuery, and Spanner with features like Lakehouse federation, Reverse ETL, Spanner Columnar Engine, and Datastream CDC, the platform aims to eliminate latency and brittle pipelines. It also expands Knowledge Catalog to provide unified governance and reduce agent hallucinations.

🚀 Google has previewed Spanner Omni, a downloadable edition of Spanner that runs outside Google Cloud — on-premises, multicloud, hybrid, and air-gapped environments. It delivers Spanner’s distributed SQL capabilities including high scalability, availability, strong consistency, and multimodal features while replacing cloud dependencies with a Colossus-like storage layer and a software TrueTime alternative. The developer preview is available for non-production use; commercial access requires engaging Google.

🚀 Small businesses are rapidly adopting Gemini Enterprise from Google Cloud to embed AI across operations, using agents to automate reporting, index internal knowledge, draft content, validate data, and streamline workflows. By making generative models accessible to nontechnical staff, the platform helps lean teams deliver faster insights and higher-quality outputs. Several SMBs worldwide report measurable productivity gains, shorter decision cycles, and reduced manual effort as they scale practical AI use cases.

🚀 Google announced a broad set of Cross-Cloud Network enhancements at Next ’26 to accelerate agentic AI, inference, and training while simplifying operations and strengthening security. Highlights include the Gemini Enterprise Agent Platform with an Agent Gateway, ambient networking for GKE and Cloud Run, and a GKE Inference Gateway for multi-region inference. The update also introduces the high-scale Virgo fabric, new Cloud Interconnect capabilities, Cloud Network Insights for observability, and expanded partner integrations and AI-driven security features.

🚀 At Next '26 Google Cloud presented a unified vision and product set to put the Agentic Enterprise into production, led by Gemini Enterprise and a new AI Hypercomputer. Announcements include the Gemini Enterprise Agent Platform and app, TPU 8-series chips for training and inference, an Agentic Data Cloud, and Agentic Defense in partnership with Wiz. Emphasis was placed on enterprise security, observability, and multi-vendor openness for regulated deployments.

🚀 Startups are accelerating agentic AI on Google Cloud, using an integrated AI stack—from models and specialized TPU/GPU compute to cross-cloud lakehouses and security—to move prototypes into production across healthcare, finance, gaming, and media. Companies like Lovable and OpenEvidence illustrate real-world adoption, while Gemini Enterprise Agent Platform, NVIDIA GPU access via the AI Hypercomputer, and Marketplace integrations aim to reduce procurement friction and speed commercialization. Google also announced a $750M partner fund and developer programs to support startups.

🚀 Google Cloud at Next '26 introduced a cross-cloud infrastructure blueprint designed for the agentic AI era, combining fluid compute, secure cross-cloud connectivity, a unified data layer, and digital sovereignty. Announcements include new CPU families (C4N, M4N with Hyperdisk Extreme), GKE Agent Sandbox, Agent Gateway, Smart Storage, Knowledge Catalog, and Confidential External Key Management to enable high-performance, governed agent workflows across clouds and on-premises. The updates target enterprises and public sector organizations preparing for machine-speed AI operations.

🛡️ Google Cloud has launched Fraud Defense, a trust platform that advances reCAPTCHA to address risks from autonomous AI agents as well as traditional bots and human fraud. The offering includes agentic activity measurement, an agentic policy engine for granular controls across the customer journey, and an AI-resistant QR code challenge to request human presence when needed. It integrates industry standards such as Web Bot Auth and SPIFEE and leverages Google’s global signals to enable largely invisible verification for legitimate users. Existing reCAPTCHA customers are automatically included with no migration or pricing changes.

🤖 Google Cloud introduced a next-generation cross-cloud Lakehouse engineered for the agentic AI era. It combines fully managed Apache Iceberg storage with read/write interoperability, a high-performance Managed Service for Apache Spark, and BigQuery integration to run multimodal workloads in real time. The service adds cross-cloud interconnect and caching to access AWS and Azure data with low latency, and unified governance via Knowledge Catalog to secure and profile data. Customers like Spotify and partners such as Accenture are already testing the platform.

🚀 Google announced a broad set of updates to Cloud Run to accelerate full‑stack app delivery, agent hosting, and high‑performance inference. New capabilities include full‑stack app deployment from AI Studio, a fully managed remote MCP server, and integration with the Gemini Enterprise Agent Platform. High‑end NVIDIA RTX PRO 6000 Blackwell GPUs are now GA, while instance primitives, SSH access, ephemeral sandboxes, and billing caps are rolling out in preview or coming soon.

🛡️ Google Cloud and Wiz outline agentic defenses, platform protections, and integrations to secure AI and multicloud workloads. At Next '26 they introduced preview and GA features including Threat Hunting, Detection Engineering, and Triage and Investigation agents, plus the Gemini Enterprise Agent Platform for agent governance. These innovations aim to automate detection, reduce analyst toil, and protect AI development and runtime across clouds and SaaS environments.

🚀 Google Cloud is expanding its partner ecosystem to accelerate the Agentic Enterprise with new funding, technologies, and integrations. The company announced a $750 million innovation fund and the Gemini Enterprise Agent Platform with an Agent Gallery to surface vetted partner-built agents. It is deepening technical alliances with global consulting firms, embedding forward deployed engineers with integrators, and extending Gemini across major SaaS platforms to speed enterprise adoption.

🤖 Google announced its AI Hypercomputer — a unified infrastructure stack built to support agentic AI — at Google Cloud Next. The announcement bundles new hardware and software, including TPU 8t and TPU 8i, A5X GPU instances, Axion N4A CPUs, the Virgo Network, and major storage and GKE upgrades. Google says the stack is designed to accelerate training and inference, reduce latency, and improve cost and energy efficiency for large-scale, agent-native applications.

🧭 Oracle AI Database@Google Cloud brings Oracle's mission-critical databases natively into Google Cloud to enable direct pipelines from enterprise records to the AI layer. The announcement expands regional availability, introduces an Oracle AI Database Agent for Gemini interaction, and integrates with Database Center, Knowledge Catalog, OCI GoldenGate, and VPC Service Controls. These features aim to lower latency, simplify governance, and make Oracle data actionable for agentic AI workflows.

🚀 Google Cloud announced a set of compute updates at Next ’26 designed to run agentic AI alongside general-purpose workloads with improved performance and lower cost. Highlights include GA for Axion N4A CPUs and GKE Agent Sandbox on Axion N4A, preview of bare-metal C4A.metal, expanded Intel Xeon 6 C4 shapes, and new high-throughput networking and Hyperdisk storage options. These changes aim to provide adaptive, secure execution sandboxes, greater I/O and network bandwidth, and flexible pricing to avoid provisioning bottlenecks and reduce TCO.

🔒 Google announced new capabilities for Google Distributed Cloud (GDC) at Next ’26, bringing Gemini models and an advanced AI stack to on-premises and edge deployments. GDC offers air-gapped and connected deployment models on Google-supplied or customer hardware, and now supports NVIDIA Blackwell GPUs, expanded machine families, and increased storage and I/O. The release adds an AI gateway for optimized inferencing — with dynamic routing, load balancing, quota controls and observability — and a sovereign agentic AI architecture on Kubernetes to run autonomous, secure agents entirely within customer boundaries.

🧭 Google announced the Agentic Data Cloud, an AI-native architecture that integrates models, analytics, and operational databases to ground agentic applications in trusted, real-time data. The release emphasizes embedding AI across the data stack, unifying transactional and analytical workloads, and simplifying enterprise deployments. New developer tools include Vibe coding integrations with Google AI Studio, modular Tools for Data Agents and onboarding/observability agents, while AlloyDB gains large-scale vector search and optimized in-database AI functions.

🚀 Google Cloud presents its eighth-generation TPUs as two specialized systems: TPU 8t for massive pre-training and embedding workloads, and TPU 8i for low-latency sampling, serving, and reasoning. TPU 8t emphasizes throughput with a SparseCore for embedding collectives, native FP4 precision, VPU/MXU overlap, and the scale-out Virgo network to reduce DCN bottlenecks. TPU 8i prioritizes on-chip SRAM, a Collectives Acceleration Engine (CAE), and the Boardfly topology to cut network diameter and tail latency. The release is paired with a performance-first AI stack — Pallas, Mosaic, native PyTorch preview, and compatibility with JAX and Pathways.