Cloud Next ‘26: Momentum and innovation at Google scale
On April 22, 2026, Google held its annual Cloud Next event, showcasing significant advancements in cloud technology and artificial intelligence (AI). Sundar Pichai, CEO of Google and Alphabet, shared insights on the rapid growth of Google Cloud and its commitment to innovation.
Key Highlights from Cloud Next ‘26
The pace of technological change has accelerated dramatically over the past year. Google Cloud is experiencing remarkable momentum, with first-party models now processing over 16 billion tokens per minute through direct API use by customers, an increase from 10 billion just last quarter. To support this growth, over half of Google’s machine learning compute investment in 2026 is allocated to the Cloud business.
Four Key Areas of Focus
During the event, Pichai emphasized four critical areas where Google Cloud is making strides:
1. The Agentic Gemini Era
Google introduced Gemini Enterprise last fall, marking a significant step into the agentic era. This end-to-end system connects data, people, and organizational goals, enabling every employee to become a builder. In the first quarter of 2026, paid monthly active users grew by 40% quarter-over-quarter, highlighting the system’s momentum.
To manage the complexity of scaling AI agents, Google is launching the Gemini Enterprise Agent Platform. This platform serves as a secure, comprehensive solution for building, scaling, governing, and optimizing AI agents, providing organizations with a mission control for their agentic enterprises.
2. AI-Driven Cybersecurity Solutions
While AI can pose security risks, Google Cloud is leveraging AI to enhance security measures for its customers. The company unveiled a new range of agentic solutions for threat detection, integrated into an AI-powered cybersecurity platform. This platform combines Google’s Threat Intelligence and Security Operations with Wiz’s Cloud and AI Security Platform.
Additionally, Google introduced Wiz’s AI Application Protection Platform (AI-APP), which offers autonomous protection across multicloud, hybrid, and AI environments, safeguarding organizations from potential threats.
3. Eighth-Generation Tensor Processing Units (TPUs)
As AI workloads become increasingly demanding, Google is evolving its infrastructure with the introduction of the eighth generation of Tensor Processing Units (TPUs). This year, Google presented a dual-chip approach:
- TPU 8t: Optimized for training, this unit scales up to 9,600 TPUs and 2 petabytes of shared, high-bandwidth memory in a single superpod, achieving three times the processing power of its predecessor, Ironwood.
- TPU 8i: Designed for inference, this unit connects 1,152 TPUs in a single pod, significantly reducing latency and delivering the throughput needed to run millions of agents concurrently at a lower cost.
These TPUs will be available to Google Cloud customers as part of the core compute processor offerings, alongside a variety of NVIDIA GPU instances.
4. Being “Customer Zero” for Innovation
Google aims to be “customer zero” for its own technologies, allowing the company to test, build, and scale its innovations effectively. This approach ensures that Google can provide the best possible solutions for its cloud customers.
Examples of this strategy include the Bigtable database service, which underpins many Google services, and the TPUs that are crucial for training and powering Gemini models. Google is also utilizing AI to generate code and address security vulnerabilities more efficiently.
Conclusion
The advancements presented at Cloud Next ‘26 illustrate Google Cloud’s commitment to driving innovation and supporting organizations in their digital transformation journeys. With a focus on AI, cybersecurity, and cutting-edge infrastructure, Google is positioning itself as a leader in the cloud computing landscape.
Note: The information presented in this article is based on the announcements made during Cloud Next ‘26 and reflects the state of Google Cloud as of April 2026.
