Google Announces Gemini Enterprise Agent Platform: The Future of Agentic AI

April 23, 20265:38264,008 views

Summary

In this keynote address at Google Cloud Next '26, Google and Alphabet CEO Sundar Pichai reflects on his 22 years at Google and outlines how the company has entered what he calls the "agentic Gemini era." Pichai frames the current moment as the fastest period of technological change he has witnessed, driven by Google's massive capital investment in AI infrastructure. He notes that Google's capex has grown nearly 6x in four years — from $31 billion in 2022 to a planned $175–185 billion in the current year — with just over half of machine learning compute in 2026 expected to benefit the cloud business.

A central theme of the talk is Google's practice of being "customer zero" for its own AI technologies. Pichai shares three internal use cases that illustrate real-world agentic AI gains. First, in software development: nearly 75% of all new code at Google is now AI-generated and engineer-approved, up from 50% the previous fall. Going further, Google deployed a multi-agent system — comprising planner, orchestrator, and coder agents — to execute a complex code migration 6x faster than would have been possible a year prior, with this approach now being extended across the full development lifecycle. Second, in marketing: when launching Gemini in Chrome, teams used AI models to generate thousands of creative asset variations, enabling personalized campaigns at scale. The result was a 70% faster turnaround time and a 20% increase in conversion rates. Third, in cybersecurity: Google's security operations center now uses AI agents to automatically triage tens of thousands of unstructured threat reports monthly — a volume that would otherwise require thousands of hours of human review. This reduced threat mitigation time by over 90%.

Pichai closes by announcing the Gemini Enterprise Agent Platform, described as "mission control for the agentic enterprise." Building on the Gemini Enterprise product launched the previous fall, this new platform provides a secure, full-stack layer for building, scaling, governing, and optimizing enterprise AI agents. He notes the conversation in the enterprise has shifted from "can we build an agent" to "how do we manage thousands of them," and positions this platform as the answer to that emerging complexity.

Key Takeaways

•Google has scaled its capex nearly 6x in four years (to $175–185B), with more than half of ML compute directed at the cloud business — signaling massive infrastructure availability for enterprise customers.
•75% of new code at Google is now AI-generated and engineer-approved, demonstrating that AI coding tools have moved well beyond novelty into standard engineering workflow.
•Multi-agent architectures (planner + orchestrator + coder) can accelerate complex code migrations by 6x — a replicable pattern for enterprise software teams.
•AI-driven creative personalization at scale produced a 70% faster campaign turnaround and 20% higher conversion for Google's own marketing teams.
•AI agents triaging unstructured threat reports cut threat mitigation time by over 90%, showing outsized impact in security operations where volume and speed matter most.
•The enterprise AI conversation has shifted from feasibility ('can we build an agent?') to management ('how do we govern thousands of agents?') — teams should plan for agent orchestration and governance infrastructure now.
•The Gemini Enterprise Agent Platform positions itself as the full-stack control plane for building, scaling, and governing enterprise AI agents.

Transcript

Thanks, Thomas. And hello, everyone. We are so glad you're here. The pace of technological change has been faster than I've ever seen. And I've been around technology for a while. In fact, this Sunday, I'm celebrating my Google vers, 22 years, and I'm still feeling lucky. And I hope that luck extends to all of you in Vegas this week. I was drawn to Google all those years ago because of its ambitious mission. From the web to mobile to AI, every platform shift has given us new opportunities to adv...