Google Cloud’s $240 Billion Backlog Signals the Agentic Shift Is Real

Google Cloud Next 2026 opened with Sundar Pichai delivering numbers that command attention. The cloud business now generates $70 billion in annual revenue, up 48% year-over-year, with a staggering $240 billion backlog—more than double last year’s $155 billion. Customers are spending 30% above their commitments, and billion-dollar deals in 2025 alone surpassed the previous three years combined. Capital expenditures are set to hit $175–185 billion this year, nearly double 2025 levels.

Pichai framed this as a fundamental rewiring of enterprise technology, positioning Google Cloud as the operating system for a new era where AI agents handle operations autonomously—chatting across platforms, managing commerce, and executing searches. The full-stack advantage is clear: custom Ironwood TPUs, frontier Gemini models, the cloud platform, Workspace, and Search reaching three billion users.

Gemini’s growth is hard to ignore. The app now has 750 million monthly active users, adding 100 million in a single quarter. AI Overviews serve two billion users across 200 countries, driving 10% more queries and appearing on 48% of them—up from 31% in February 2025. The Gemini API handled 85 billion requests in January 2026, a 142% jump from March 2025. Eight million paid Gemini Enterprise seats are deployed across 2,800 companies, and 13 million developers are building on generative models. Gemini 3 Pro is seeing the fastest adoption of any Google model.

Search itself is evolving from a retrieval tool into an agent manager, dispatching tasks across multiple threads. Pichai envisions a future where shopping agents bridge discovery and purchase, eliminating indecision. To enable this, Google launched the Universal Commerce Protocol (UCP), an open-source standard co-built with Shopify, Etsy, Wayfair, Target, and Walmart, and backed by Adyen, American Express, Best Buy, Flipkart, Macy’s, Mastercard, Stripe, Home Depot, Visa, and Zalando. UCP runs on REST and JSON-RPC, integrating with the Agent2Agent protocol, Model Context Protocol, and Agent Payments Protocol. Stores become programmable services; agents buy on your behalf.

The Agent2Agent protocol has matured to version 1.2, now governed by the Linux Foundation’s Agentic AI Foundation and in production at 150 organizations—no pilots. Microsoft, AWS, Salesforce, SAP, and ServiceNow all run it. A2A handles cross-boundary agent communication, complementing Anthropic’s MCP for tool connections. Google also rolled out managed MCP servers for Maps, BigQuery, Compute, and Kubernetes, while Apigee bridges APIs to agent tools.

Thomas Kurian, Google Cloud CEO, took a swipe at competitors: “They hand you pieces, not the platform.” Google’s stack includes eighth-generation TPUs split between training (Broadcom) and inference (MediaTek), the Gemini 3 family, and the Gemini Enterprise Agent Platform (formerly Vertex AI). Workspace Studio enables no-code agent creation. By contrast, Microsoft Copilot has only 3.3% paid adoption in Microsoft 365 and a negative Net Promoter Score. Google’s eight million seats are growing faster.

New TPUs target agentic workloads directly. The TPU 8t delivers 3x faster training with 9,600 chips per superpod and 121 exaflops of performance. The TPU 8i focuses on inference. The Virgo Network links up to 134,000 TPUs at 47 petabits per second. Google also announced the A5X with NVIDIA Vera Rubin, cutting inference costs by 10x and boosting throughput per megawatt. NVIDIA named Google Cloud its AI Global Tech Partner and Infra Modernization Compute winner at Next.

A $750 million agentic AI partner fund was launched, targeting everything from value identification to deployment and upskilling. Accenture and KPMG are early participants. KPMG achieved 90% Gemini Enterprise adoption in a month. GE Appliances runs over 800 agents in manufacturing. ServiceNow signed a $1.2 billion, five-year deal. Cognizant launched Agentic Retail CX on Gemini for omnichannel retail, and Salesforce integrated for cross-platform agents in Slack and Workspace.

Security is getting agentic too. Threat Hunting and Detection Engineering agents probe malware and auto-generate rules. Wiz AI Application Protection scans code from development to runtime across multi-cloud environments. Google Unified Security ties it together. Agent Gateway enforces policies on MCP and A2A traffic, while Agent Identity uses cryptography for traceable authentication. Model Armor blocks prompt injection and data leaks.

Internally, 75% of Google’s new code is AI-generated, up from 25% in late 2024 and 50% last fall. Agents sped a code migration sixfold. Engineers still review; some teams use Claude Code. Externally, the Gemini Enterprise Agent Platform unifies low-code agent creation, orchestration, registry, and observability via OpenTelemetry. Project Mariner browses the web, and Workspace Intelligence fuses Gmail, Docs, and Drive for automated reports and actions.

Capital expenditure breaks down as 60% servers and 40% data centers and networking. Big tech AI spending is approaching $700 billion. Constraints on power, land, and supply chains will persist through 2026. Google Cloud holds 11% market share, trailing AWS (31%) and Azure (25%), but its 48% Q4 growth and profitability are standouts. The backlog covers more than three years of revenue, and 13 product lines each exceed $1 billion.

The Agentic Data Cloud shifts data from a repository to a reasoning engine. Cross-Cloud Lakehouse queries AWS and Azure data via Iceberg without moving it. Knowledge Catalog and Smart Storage extract context from PDFs and images. C4N and M4N VMs handle agent traffic, with fluid compute across Engine and Kubernetes.

That 75% AI-written code stat lingers. Humans check the output, but as agents scale, review bottlenecks loom. Google is betting its integrated stack—from custom silicon to the inbox—captures value beyond market share. Rivals fragment. Google integrates. The race isn’t just about models anymore. It’s about agents, interoperability, governance, and production at scale.

Pichai’s closing vision: Search as conductor, agents as orchestra, enterprises transformed. Supply is tight. Demand is exploding. Google Cloud is riding the wave.

Source: Webpronews