All news with #google cloud tag

🔒 A Qualys survey and analysis of roughly 44 million public-cloud VMs highlights widespread misconfiguration: 45% of AWS, 63% of GCP and 70% of Azure instances showed issues. Respondents reported breaches and identified misconfigured services as a leading cloud risk. Experts cite neglected logging, monitoring and MFA, rushed M&A integrations and understaffed small firms as common causes. The piece recommends concrete controls — from Infrastructure as Code and continuous scanning to private networking and least-privilege — to reduce exposure.

📧 Check Point disclosed a large-scale phishing campaign that abused Google Cloud Application Integration to send authentic-looking messages from noreply-application-integration@google[.]com, enabling attackers to bypass SPF and DMARC protections. The emails mimicked routine enterprise notifications to prompt clicks and redirected victims through Google Cloud storage to a fake CAPTCHA and a counterfeit Microsoft login page. Google has blocked the abuse and is implementing further mitigations.

📢Google Cloud’s Data Cloud updates through mid‑2025 introduce new self‑service Looker features, expanded Model Context Protocol (MCP) support, and tighter AI-to-data integrations. Highlights include AlloyDB AI time‑series forecasting via AI.FORECAST, GA of Conversational Analytics powered by Gemini, and the MCP Toolbox and ADK to securely connect agents to BigQuery, Spanner, Cloud SQL, and Looker. Dataplex Universal Catalog now previews curated data products for governed, deployable datasets and AI use.

🔒 Since the 2017 breach, Equifax has pursued a comprehensive security transformation, investing nearly $3 billion to rebuild technology and migrate to Google Cloud under NIST-aligned frameworks. The company reports that security is now embedded across processes and incentivized through employee bonuses, with regional CISOs adapting programs to EU rules like DORA and NIS2. Equifax says it neutralizes millions of threats daily and uses a hybrid approach to AI-driven attacks, combining multiple layers of controls rather than relying on a single technology.

🛡️Attackers abused Google Cloud Application Integration to send thousands of malicious emails that appeared to originate from the legitimate address noreply-application-integration@google.com. The messages impersonated routine enterprise notifications—voicemail alerts, file-access and permission requests—raising the chance recipients would click links or disclose credentials. Check Point observed 9,394 phishing emails targeting about 3,200 customers over 14 days.

📅 Google Cloud’s 2025 roundup highlights a year of fast-moving AI and infrastructure innovation, from major model releases to accelerator and networking expansions. The blog chronicles product launches like Gemini 3, Veo 3, Ironwood TPUs, and Vertex AI enhancements alongside security work, research, and high-profile customer stories. It traces month-by-month developments, acquisitions and partnerships, and guidance for enterprises adopting agentic AI and cloud-scale systems.

🔍 Looker is expanding self-service analytics with three Public Preview capabilities: self-service Explores, tabbed dashboards, and internal dashboard theming. Users can upload CSVs and spreadsheets, join local files with governed models, and visually create measures and dimensions while administrators control upload permissions and certify content. Tabbed dashboards organize complex narratives with cross-tab filters and shareable tab URLs, and internal themes let teams apply consistent styling across dashboards.

🔒 Google Cloud senior leaders Nick Godfrey and Anton Chuvakin recap 2025 security developments and lessons learned across cloud and AI. They highlight five focus areas — securing cloud, securing AI, AI-enabled defense, threat intelligence, and building trust — and call out major items such as the announced Wiz acquisition, response to React2Shell (CVE-2025-55182), and the launch of AI Protection. The newsletter stresses fundamentals, governance, and using AI to empower defenders.

🔬 Stochastic rounding restores tiny gradient updates that deterministic low-precision formats would otherwise zero out, enabling stable training in FP8 and 4‑bit regimes. Frameworks such as JAX and the Qwix quantization toolkit apply SR on Google Cloud accelerators—TPU MXUs and NVIDIA Blackwell A4X VMs—to prevent vanishing updates. The approach trades deterministic bias for unbiased noise, often acting as implicit regularization and preserving model convergence while boosting efficiency.

🚀 These hands-on labs guide developers through three Google Cloud deployment options to move AI agents from local prototypes to production. The Vertex AI Agent Engine offers a fully managed, Python-optimized runtime that handles execution, memory, and tool invocation. Cloud Run provides a serverless container experience with autoscaling and language flexibility, while GKE delivers orchestrated control for microservice deployments.

🔬 Google is partnering with the DOE's Genesis Mission to accelerate federally funded scientific discovery by combining high-performance computing, experimental facilities, and AI. Gemini for Government and Google DeepMind tools offer multimodal reasoning, agentic workflows, and an AI co‑scientist to speed hypothesis development. Google Cloud provides the secure, accredited infrastructure for multi‑lab deployments.

🔍 Google Cloud describes how the AlloyDB AI natural language API translates user questions into SQL and how to tune it for near‑perfect accuracy in enterprise applications. The post outlines a hill‑climbing workflow that improves results by adding descriptive (table and column) and prescriptive (templates, facets) context, plus an automated value index for private terms. It highlights capabilities for business relevance, explainability, and verified results, and explains agent integration options such as the MCP Toolbox and Gemini Enterprise.

🔒 The Zeroday Cloud competition in London, hosted by Wiz Research with support from AWS, Microsoft, and Google Cloud, awarded $320,000 to teams that demonstrated 11 zero-day remote code execution vulnerabilities. Exploits affected critical cloud components including Redis, PostgreSQL, MariaDB, Grafana, and a Linux-kernel container escape that broke tenant isolation. Team Xint Code earned the top prize of $90,000. Attempts against AI tooling such as vLLM and Ollama were made but failed due to time exhaustion.

🚀 Cluster Director is now generally available on Google Cloud, offering a managed, topology‑aware control plane for Slurm clusters to streamline AI and HPC operations. It automates Day 0 planning, Day 1 deployment, and Day 2 monitoring with health checks, Managed Lustre attachment, topology‑aware placement, and one‑click node remediation. There is no additional fee for Cluster Director — customers pay only for underlying Google Cloud resources. Google also announced a Preview of Slurm on GKE to combine Slurm's batch scheduling with Kubernetes' autoscaling and self‑healing.

🏆 Google Cloud was named a Leader in The Forrester Wave™: AI Infrastructure Solutions, Q4 2025, earning the top score for Current Offering and the highest possible marks in 16 of 19 evaluation criteria. The post emphasizes an integrated AI Hypercomputer that co‑designs custom TPUs with NVIDIA GPUs and a high‑bandwidth network fabric to optimize throughput, latency, and TCO. It highlights operational tools like GKE Autopilot and Cluster Director, cost controls such as Dynamic Workload Scheduler, and security investments including Titan chips and Security Command Center.

🚀 The new Google Cloud Partner Network is a restructured program launching in Q1 2026 that shifts partner evaluation from checklist activity to measurable customer outcomes. It centers on three pillars — simplicity, outcomes, and automation — and introduces a three-tier model: Select, Premier, and a selective Diamond tier. A competency framework measuring capacity and capability, together with an automated Partner Network Hub, will reduce administrative burden and automatically credit qualifying customer engagements.

🚨 Google Workspace with Gemini helps agencies convert complex, legally precise emergency documents into clear, actionable public-facing tools almost instantly. By parsing executive orders to extract locations, curfews, and exemptions and generating interactive prototypes or safety-check interfaces directly in the conversation window, the system shortens deployment from days to minutes. Built with public-sector security like FedRAMP High, data residency controls, DLP, and client-side encryption, it provides a repeatable blueprint for rapid, trustworthy disaster response.

🔌 Google Cloud has integrated the Model Context Protocol (MCP) into Antigravity, its new AI-first IDE, enabling LLM-based agents to access enterprise data services directly within the development workflow. The Antigravity MCP Store lets developers install connectors for AlloyDB, BigQuery, Spanner, Cloud SQL, Looker and other Data Cloud products, configuring projects, regions, and credentials through a guided UI. Once connected, agents receive executable tools for schema exploration, query development, optimization, forecasting, catalog search, and semantic validation, while credentials are stored securely and MCP standardizes access across services.

📈 This article describes a multi-agent business forecasting application developed by Google Cloud and App Orchid. The design pairs a Google prediction agent (leveraging TimesFM and the Population Dynamics Foundation Model) with an App Orchid Data Agent that builds a semantic knowledge graph and prepares AI-ready time-series. A forecasting orchestrator uses the A2A Protocol and Google’s ADK to route queries, automate data wrangling, run predictions on Gemini-powered Vertex AI, and return unified forecasts with enterprise-grade security and governance.

🔍 IDC warns of a growing Total Cost of Ownership (TCO) crisis as AI inference becomes the dominant workload. Their global survey of 1,300 AI decision-makers finds inference already accounts for 47% of AI operations and is magnified by agentic workflows that trigger many sequential model calls. The research attributes the problem to fragmented stacks and idle accelerators and recommends shifting to integrated, system-level architectures that unite software, storage, networking, and compute. Google Cloud highlights AI Hypercomputer as a purpose-built solution to improve utilization and cost-effectiveness.