Ordnance Survey turn from Cartography to Code through harnessing AI
Since joining Ordnance Survey (OS), I’ve had the privilege of being part of an organization undergoing one of the most significant transformations in its history. Of course, when you think of OS, you rightly think of maps. But do you also think of machine learning and being at the forefront of AI, computer vision and tech innovation?
You’d be forgiven if not. After all, OS was first founded back in 1791, and since then has undergone huge transformation from its heritage in paper cartography to where we are today with over 500 million features being updated 20,000 times a day.
Of course, that level of data analysis can’t be achieved by the human eye alone. Here at OS we’ve been embedding AI into the way we collect and modernize geospatial intelligence for well over a decade. For us, AI isn’t a buzzword. It’s the most powerful and practical tool that enhances our mapping capabilities and enables us to deliver faster, smarter insights to our customers and partners.
Now, we’re entering a new era at OS, one defined by agentic AI, where we’re integrating large language models (LLMs) with more advanced, autonomous agents capable of executing increasingly complex tasks. This innovation comes with its challenges, and we’re working to ensure we continue to provide location data responsibly, while taking advantage of the power that AI tools provide.So how is OS currently using AI?
At a foundational level, we’re leveraging the power of AI to automatically extract features from our aerial and street-level imagery, enabling us to detect changes and update our maps rapidly and at scale.
This allows for near real-time updates that identify shifts in land use, infrastructure, and buildings — supporting local authorities and businesses with planning, compliance, urban development, and much more.
