Google Cloud Next 2025: The Announcements You Need to Know

Google has presented its latest vision for enterprise AI at Google Cloud Next in Las Vegas this week, making a series of hardware and software announcements as its major cloud rivals continue to compete for AI market share.

The event comes at a pivotal moment for the cloud computing industry. Google, along with its cloud giant rivals Amazon Web Services (AWS) and Microsoft, are investing heavily in AI infrastructure as they position themselves for what they see as the next wave of enterprise technology adoption.

In a blog post published ahead of the conference, Google Cloud CEO Thomas Kurian set expectations for the event. “Today, that vision is not just a possibility – it’s the vibrant reality we are collectively building,” he wrote.

In a year that’s seen AI stocks stumble and tech giants face growing scrutiny over their infrastructure investments, Google has highlighted several metrics, including “3,000 product advancements” across Google Cloud and Workspace in 2024, and a “20x increase in Vertex AI usage” over the past year. Google also claimed its platform now supports more than four million developers building with its Gemini model family.

Google AI hypercomputer strategy takes shape with Ironwood TPUs

Google has unveiled its seventh-generation Tensor Processing Unit (TPU) named Ironwood, challenging Microsoft Azure and AWS in the enterprise AI hardware market. The announcement was part of a broader set of updates aimed at organisations implementing AI systems.

The Ironwood TPU provides 42.5 exaflops of compute per pod with over 9,000 chips. Google described it as “more than 10x improvement” compared to its previous TPU generation.

Google also says its AI Hypercomputer system delivers “24x higher intelligence per dollar compared to GPT-4o and 5x higher than DeepSeek-R1” when running Gemini 2.0 Flash.

“Our AI Hypercomputer is a revolutionary supercomputing system meticulously designed to simplify AI deployment, dramatically improve performance and optimise costs,” Thomas writes. “It includes hardware, software and consumption models – all optimised to deliver more intelligence at a consistently low price for training, tuning and serving AI workloads.”

Google Cloud Wide Area Network opens network infrastructure to corporate clients

Google is also announcing it will offer its global private network to businesses through a new Cloud Wide Area Network (Cloud WAN) service: reportedly providing “up to 40% improvement in network performance while reducing total cost of ownership by a similar percentage.”

The move puts Google in competition with both telecom providers and cloud rivals by making its infrastructure – comprising over two million miles of terrestrial and subsea cables with 200+ points of presence worldwide – available to organisations looking to modernise their networks.

This development comes alongside Google’s continued data centre expansion, which now includes 42 cloud regions. Recent additions include facilities in Sweden, South Africa and Mexico, with projects in Kuwait, Malaysia and Thailand underway. The geographic spread addresses both regulatory requirements for local data storage and technical needs for lower latency in AI applications.

“Starting today, this network, which moves at ‘Google speed’ – near-zero latency – for billions of users worldwide, is now available to enterprises everywhere,” said Thomas.

Gemini 2.5 Flash targets cost management in enterprise AI deployment

Google also recently introduced Gemini 2.5 Flash, a model that focuses on balancing speed with operating costs – a key concern for businesses implementing AI at scale. At Next, the company said the model will be available through the Vertex AI platform and can adjust processing depth according to query complexity.

“Flash is ideal for everyday use cases like providing fast responses during high-volume customer interactions, where real-time summaries or quick access to documents are needed,” said Thomas. “Gemini 2.5 Flash adjusts the depth of reasoning based on the complexity of prompts, and you can control performance based on customers’ budgets.”

This release comes two weeks after Google began a public preview of Gemini 2.5 Pro. The company positions Pro for technical tasks like code development or healthcare document analysis, while Flash targets routine business processes where cost control matters more than computational depth.

Google also updated its media generation tools. Imagen 3, the text-to-image platform, received enhanced inpainting functions for repairing images, while Chirp 3 – for audio generation – can now create voice replicas from 10-second samples. Veo 2, the video generation tool, added capabilities including inpainting, outpainting and new cinematic techniques.

Google shifts focus to multi-agent AI systems

Google has also launched an Agent Development Kit (ADK), with the open-source framework allowing developers to create an AI agent with “under 100 lines of code.”

The company also introduced an Agent2Agent (A2A) protocol for inter-agent communication regardless of underlying technology. Google mentioned that 50+ companies – including Accenture, Box, Deloitte, Salesforce, SAP, ServiceNow and TCS – are working on this protocol standard.

“We believe Vertex is the most open developer AI platform in the cloud and the only one delivering multi-agent solutions – empowering multiple AI agents to work together,” said Thomas.

On the security front, Google announced Google Unified Security, which includes “visibility, threat detection, AI-powered security operations, continuous virtual red-teaming, enterprise browser security and Mandiant expertise.” New capabilities include alert triage for security investigation and malware analysis for examining suspicious code.

For organisations with strict data requirements, Google formed a partnership with Nvidia to run Gemini on Nvidia Blackwell systems, with Dell as a hardware partner. This enables on-premises deployment in regulated environments, potentially addressing adoption barriers in healthcare, finance and government sectors.

For organisations with strict data requirements, Google has formed a partnership with Nvidia to run its models to on-premises environments, with Dell as a hardware partner. This enables on-premises deployment in regulated environments, potentially addressing adoption barriers in healthcare, finance, and government sectors.

“Nvidia and Google Distributed Cloud provide a secure AI platform, bringing Gemini models to enterprise data centres and regulated industries,” explains Justin Boitano, VP, Enterprise AI Software, Nvidia. “With Nvidia Blackwell infrastructure and confidential computing, Google Distributed Cloud enhances privacy and security, and delivers industry-leading performance on DGX B200 and HGX B200 systems, available from Dell.”

“The opportunity presented by AI is unlike anything we've ever witnessed,” Thomas concludes. “It holds the power to improve lives, enhance productivity and reimagine processes on a scale previously unimaginable.”

Explore the latest edition of Technology Magazine and be part of the conversation at our global conference series, Tech & AI LIVE.

Discover all our upcoming events and secure your tickets today.

Technology Magazine is a BizClik brand