June 17, 2026, (Inside AI) — The UK government announced a national rollout of its Extract tool and progress on the Augmented Planning Decisions prototype at the Google Cloud Summit London. Both projects aim to modernize local planning systems using Google Cloud infrastructure.
The Blueprint for Digital Planning Takes Shape
The Ministry of Housing, Communities and Local Government (MHCLG) and the Department for Science, Innovation and Technology (DSIT) revealed that Extract is now available to all councils in England. The tool automates document processing, saving an estimated 255 hours of manual work per council.
Meanwhile, the Augmented Planning Decisions (APD) prototype is in alpha testing with three local authorities. It assists planning officers in navigating complex policies. A national release is targeted for 2027.
Both tools run on Google Cloud, but their origins differ. MHCLG and the Incubator for AI (i.AI) built Extract internally. The APD prototype involves Google Cloud, Google DeepMind, and Faculty.
Read: AI Adoption in UK Reaches Tipping Point, Google Cloud Exec Says
Why Cloud Infrastructure Underpins the Strategy
Scaling AI across 300+ local authorities demands elastic, secure infrastructure. Google Cloud provides the foundation to move from experiment to production. The government retains full ownership of the digital transformation.
Debbie Weinstein, Vice President at Google Cloud, stated the platform’s role is to support, not drive, the vision. She emphasized that the government is the primary architect, while Google Cloud ensures scalability and reliability.
Security and Sovereignty in the Spotlight
Handling sensitive planning data at national scale requires rigorous risk management. Extract uses Gemini on Google Cloud, keeping data within a protected environment. This mitigates threats like prompt injection and upholds data sovereignty.
Weinstein noted that the approach allows the government to use advanced AI reasoning securely. The infrastructure is designed for enterprise-grade demands from day one.
Collaboration Without Vendor Lock-In
The ecosystem combines MHCLG’s policy expertise, i.AI’s technical skills, DeepMind’s research, and Faculty’s delivery. Google Cloud’s role is foundational, not proprietary. The government could choose alternative platforms for later stages.
This model counters narratives of big tech overreach. The tools remain government-led, with cloud services as interchangeable components. It’s a pragmatic balance between innovation and control.
Read: Maharashtra Inks Free AI Training Deal with Google for 400,000 Teachers
From Alpha to National Impact
The APD prototype is still early and experimental. Results are not yet conclusive. If successful, Google Cloud is positioned to scale the service seamlessly. The 2027 target depends on continued testing and policy alignment.
Extract’s rollout proves the model works. Its nationwide availability signals confidence in cloud-based AI for public services. The government expects more efficient, transparent planning systems as a result.
Weinstein concluded that the collaboration demonstrates how public sector vision supported by enterprise technology yields practical outcomes. The focus remains on production-ready tools, not just pilots.
