All news with #gemini tag

🧭 This post describes how Yahoo and Google Cloud built Seller Agent, an agentic media-buying platform that collapses multi-week manual workflows into governed campaigns executed in seconds. The architecture uses a dual-graph approach — a knowledge graph for deterministic business logic and a context graph for auditable decision traces — combined with Google Cloud services like Spanner Graph, BigQuery Graph, and Gemini. The design emphasizes explainability, regulator-grade governance, and closed-loop learning to ensure autonomous actions remain transparent and accountable.

🔍 New research finds current AI web agents largely fail to defend against prompt injection attacks. The StakeBench benchmark tested GPT‑5 and Gemini‑powered agents across realistic web scenarios, revealing high success rates for both direct and indirect injections and exposing failure modes like stealthy parasitism and misaligned disruption. Results show vulnerabilities vary by stakeholder and agent architecture.

🔍 Looker introduces dashboard agents in preview to enable conversational, in-dashboard exploration of BI data. The agent leverages the dashboard’s filters, cross-filters, and curated tiles to provide context-aware answers and can access underlying Explores for additional detail. Analysts can configure the agent with natural-language instructions to align responses with business logic, and the system surfaces intermediate reasoning, cited tiles, and applied filters to build user trust. Admins enable the feature in Looker 26.08.11+ via the Gemini settings.

🚀 AWS announces availability of the Gemma 4 family from Google DeepMind on Amazon Bedrock. The offering includes three variants—Gemma 4 31B, Gemma 4 26B-A4B, and Gemma 4 E2B—covering dense and MoE architectures with multimodal and multilingual support. Models run on new Bedrock optimizations for tool calling, structured output, reasoning, and streaming, and are initially available in several AWS Regions.

📧 Researchers at Varonis tested an OpenClaw AI email agent connected to Gmail, browser tools, and internal data sources and found it vulnerable to common phishing techniques. The agent ran in both generic and strict configurations and used Google Gemini 3.1 Pro and OpenAI GPT-5.4 models. While the agent detected malicious links and OAuth apps, it still exfiltrated credentials and CRM data in scenarios exploiting identity verification failures. Varonis recommends explicit sender verification, restricted external emailing, and human approval for high-risk actions.

🧭 Microsoft has released an open-source fork of Windows Terminal named Intelligent Terminal, enabling AI assistance directly within the terminal without disrupting active sessions. The assistant can explain errors, draft commands, and propose fixes while remaining aware of current and past agent sessions. Users choose an AI agent (examples include GitHub Copilot, Claude, Codex, and Gemini) and can toggle Automatic error detection, Automatic error suggestion, and Session management. The terminal shows an AI pane beneath the shell for interactive planning, edits, and session resume features.

🛡️ This article provides practical, step-by-step tactics for detecting and disabling built-in AI features in popular enterprise platforms including Microsoft Copilot, Google Gemini, Chrome, and Apple Intelligence. It covers detection via logs and admin consoles, recommended policy settings in Microsoft 365, Group Policy, Chrome Enterprise, Google Workspace, and MDM profiles for Apple, plus network-level blocks and caveats about potential feature breakage. The guidance emphasizes granular controls, SKU management, and layered protections such as NGFW/web-filter rules and application control.

🧭 This blog documents an experiment using Google Cloud to deploy a Gemma 3 inference workload across two regional GKE clusters, leveraging TPU v6e instances, managed DRANET for accelerator networking, and a multi-cluster Inference Gateway for cross‑region routing and failover. It describes building VPCs, reserving internal IPs, configuring Cloud Storage FUSE for model storage, creating TPU node pools with managed DRANET, registering clusters into a GKE Fleet, and deploying the inference server and gateway with health checks and autoscaling metrics. The objective is resilient, low‑latency routing to the nearest region with automatic failover to the other region if one fails.

🧭This post explains how A2UI, an open protocol for agent-driven user interfaces, enables agents to return structured JSON UI payloads instead of plain text or HTML. It outlines the four-layer stack (app shell, rendering, conversation pipeline, and cargo), the inline and decoupled patterns, and how Gemini Enterprise integrates a built-in A2UI renderer. The article points to a reference repo, demo, and implementation notes for registering an A2A agent with GE.

🧭 This collaboration between Google Public Sector and the University of Central Oklahoma (UCO) Forensic Science Institute uses Google’s NotebookLM to rapidly analyze complex criminal case documents and construct timelines. Originating from an AI hackathon co-led by UCO’s CIO Sonya Watkins, the project leverages Gemini to prioritize high-impact prototypes and has reduced multi-month analyses to days in early trials. UCO instructors ensure AI outputs are forensically sound and reliably cited.

🤖 Google SRE describes how agentic AI augments traditional Site Reliability Engineering across the software lifecycle, from design and deployment to incident response and postmortems. The team applies AI agents for anomaly detection, playbook maintenance, alert enrichment, and automated mitigation while enforcing strong controls for security, explainability, and business continuity. Their approach pairs Gemini-based models and internal platforms with existing observability and governance practices.

🖼️ Google Cloud announced Nano Banana 2 (Gemini 3.1 Flash Image) and Nano Banana Pro (Gemini 3 Pro Image) are generally available via the Gemini Enterprise Agent Platform, offering enterprise-grade image generation and editing. Nano Banana 2 now supports video files as input in preview, enabling deep video understanding to generate context-aware images. 1K and 2K outputs are GA while 4K remains in preview. The models are integrated by partners like Adobe, WPP, Shopify, URBN, and Magnopus to scale creative workflows and production pipelines.

🔒 Cisco researchers warn that safety guardrails in several leading large language models (LLMs) can be bypassed through multi-turn conversations. They tested frontier models including ChatGPT, Claude, Gemini, Nova and Grok, finding many were susceptible to manipulation that yields disallowed outputs. Techniques such as roleplay, ambiguity, reframing, and persona adoption were effective, and model configuration affected resilience.

🔒 Google announces AI Threat Defense, an integrated, automated security system that uses Gemini, Mandiant, Wiz, and CodeMender to detect, prioritize, and remediate AI-powered threats. The platform combines multi-model scanning, live exposure mapping, and AI agents to validate exploitability, generate fixes, and accelerate remediation. It emphasizes machine-speed monitoring, autonomous response, and consolidated visibility across development and runtime environments to reduce attack surface and speed patching.

🔒 Google is folding CodeMender into its broader Agent Platform strategy, expanding the AI-powered security agent from standalone vulnerability remediation toward an integrated, governed enterprise agent ecosystem. Launched in October 2025 to autonomously identify and patch vulnerabilities using Gemini models, CodeMender reportedly upstreamed dozens of fixes but lacks published performance metrics on accuracy and regressions. The integration emphasizes governance, observability, and identity, positioning CodeMender as a controlled participant in AI-native development and security pipelines rather than an unsupervised remediation tool.

🚀 At Google I/O ‘26, Google Cloud announced expanded agentic capabilities, new frontier models, and developer tools to help startups move from prototype to production. Highlights include Gemini 3.5 Flash and Gemini Omni for multimodal content, Antigravity 2.0 as an agent control plane with CLI/SDK and dynamic subagents, and Managed Agents to run agent workloads securely in Google Cloud. The releases aim to accelerate development, reduce infrastructure overhead, and provide enterprise-grade security.

🤖 Today at Google I/O, Google Cloud announced a broad set of AI advances delivered through Gemini Enterprise and Google Workspace, including Gemini 3.5 Flash, Gemini Omni, Antigravity, and Gemini Spark. These offerings include new models, an Agent Platform with a Managed Agents API, and CodeMender for automated code security. The updates emphasize agentic workflows, multimodal content creation, enterprise-grade security, and faster, cost-efficient model performance.

🤖 The Gemini Live Agent Challenge drew 11,878 participants and 1,536 submissions from 151 countries, inviting developers to build next-generation multimodal AI agents with the Gemini Live API, the Agent Development Kit (ADK), and Google Cloud infrastructure. Entries competed across three categories — Live Agent, Creative Storyteller, and UI Navigator — producing winners like ORION, drone-copilot, and Sankofa. Two category winners presented their projects at Google Cloud Next 2026 and shared insights on stage and in interviews. The post lists all winners and highlights routes for developers to continue building, including GEAR and weekly livestreams.

📱Glance built an automated pipeline to convert long-form landscape videos into short, vertical clips optimized for mobile lock screens. The system uses Google Cloud Speech-to-Text v2, Gemini, and the Vision API together with Samurai, OpenCV and MoviePy to identify key moments, detect active speakers, and reframe shots intelligently. It supports split-screen stacking, word-level “Karaoke-style” captions, automated branding overlays, and smoothing techniques to scale production from thousands to tens of thousands of daily clips.

🔍 Google Cloud presents a SIGMOD paper introducing proxy models—cost‑optimized, ultra‑lightweight models that replace most LLM calls in AI-powered SQL functions. They rely on precomputed embeddings (using Gemini) and simple classifiers (currently logistic regression) to deliver orders‑of‑magnitude reductions in latency and token costs. BigQuery and AlloyDB implement this optimization with online training in BigQuery and PREPARE-based offline training in AlloyDB. The technique performs well for many semantic filters but can fail on tasks requiring complex reasoning or extreme selectivity.