Summary Bullets:

• The Google Cloud UK Summit was held in June 2026, at Tobacco Docks, in London (England). The company will release Gemini 3.5 Pro and Gemini Omni, a multi-modal system, in late-2026.

• Google announced the UK availability of Gemini 3.5 Flash, through the Gemini Enterprise Agent Platform and the Gemini Enterprise app.

Google Cloud held its UK Summit in London with the strategic intent of demonstrating the ROI of agentic AI. With more than 4,000 participants, it followed the thread commenced in 2025 (please see Google Cloud Focuses on Agentic AI During UK Summit, July 15, 2025), only this time the industry has moved decisively towards the implementation of the technology in practical cases. The company showcased its considerable presence in the UK, with a new data center that will be opening in Waltham Cross, Hertfordshire, later in 2026 as part of a GBP5-billion investment in AI programs in the country.

Google Cloud is collaborating with the Ministry of Housing, Communities and Local Government (MHCLG), and the Department for Science, Innovation, and Technology (DSIT) in the launch of a prototype AI tool for urban planning. This prototype is currently being tested and developed by three local authorities in London, with the potential to be deployed over 300 more across the UK. If MHCLG chooses, Google Cloud could benefit from the contract to roll out the service once it is made available to councils nationally from 2027.

The AI juggernaut emphasized the business value of agentic applications in production environments. An example of this was the Extract AI tool (developed internally by MHCLG and the Incubator for AI (i.AI)), a tool that automates the processing of unstructured data into digitalized geospatial information in structured formats, leveraging LLMs including Google DeepMind’s Gemini, and Computer vision technology by Meta’s Segment Anything Model. Extract AI has been rolled out to all English councils and outcomes have been estimated to yield savings of 255 hours of manual work per site.

In the private domain, Google Cloud showcased collaborations with Unilever, THG Ingenuity, Rightmove, Kingfisher, and more. It also announced a partnership with Deloitte, which has launched an AI studio on its London campus, to deliver agentic AI projects to enterprises in the UK. Opening in late July, it focuses on industry-specific solutions to help organizations with agentic deployments. The professional services company will also upskill 1,000 of its technology specialists on Google Cloud’s Gemini Enterprise to implement Google’s latest agentic architecture to validate complex projects in only four weeks.

Google Cloud also announced a partnership with HSBC, leveraging agentic AI to enhance personalized wealth management, financial crime risk, and manager client services. The bank is working with the Google DeepMind engineering teams to collaborate on AI initiatives running on Gemini Enterprise Agent Platform. The project will enable 200 use cases over two years, focusing on deployments with an estimated benefit quantifiable at over $100 million.

Google Cloud’s offerings, many released during Google I/O 2026, held in Mountain View (CA), last May, support a strategic intent to embed agentic AI across the enterprise, from software development to line-of-business applications. The Gemini Enterprise Agent Platform acts as the centralized control layer for coordinating autonomous processes, supported by Agent Designer v2, a low-code tool for citizen developers to create agents using natural language. For engineering teams, Antigravity 2.0 applies the same agent-first approach to coding. In the realm of personal productivity tools, Gemini Spark is an agentic personal assistant, running on Gemini 3.5 Flash, whose roadmap features will be expanded in H2 2026 to include integrations with third-party apps using Model Context Protocol (MCP).

During the event, the company argued that agentic implementations rely not only on powerful agentic capabilities but careful and methodical processes. Governability, observability and security will “make or break” the success of live deployments. However, this is such a dynamic and competitive segment of the market that demonstrating deep infrastructure reach, enterprise workflow distribution, and ecosystem gravity will prove the ultimate test for Google Cloud. Solving these challenges will decide who comes out victorious in the agentic AI arms race.