Google just made Google Earth way more fun for all of us! You can now generate AI images directly on top of real‑world locations using Nano Banana 2. Zoom to a place, hit “create image,” type a prompt – and Earth transforms into a creative canvas for maps, cities, and futures.
What makes this #geoawesome is the combination of grounded geography and wild imagination. You’re not starting from a blank canvas: you’re remixing satellite, aerial and 3D Earth scenes into historic reconstructions, speculative futures, and playful urban utopias.
Of course, my first instinct was to see if I could get Earth + Nano Banana to spell “GEOAWESOME” directly from a satellite view – turning buildings, streets and green spaces into giant letterforms for our community name.
Right after that, I jumped into a Berlin future vision:
Transform central Berlin into a luminous mobility utopia with glowing elevated U‑Bahn tracks, flying transport pods gliding between stations and cascading ivy and wild vines wrapped around buildings, ultra‑detailed, dusk lighting, viewed from a high oblique angle.
For the Geoawesome community, this unlocks rapid visual sketching for planning workshops, climate adaptation stories, education, and just pure creative exploration. You can prototype ideas in context, share before/after views, and let people literally see what “what if?” looks like at the neighborhood or city scale.
In short: Google Earth is no longer just a place to look at the world, it’s becoming a place to reimagine it. That’s peak #geoawesome.
You can read Google’s full announcement and get started with Earth + Nano Banana here:
Transform any place with Nano Banana in Google Earth
#Fun
Next article
HERE Technologies and Esri have announced a three-year agreement to jointly develop location analytics and visualization capabilities for artificial intelligence applications.
Unveiled at the 2026 Esri User Conference, the agreement combines HERE’s mapping, traffic and mobility data with Esri’s ArcGIS platform. The companies say the resulting capabilities will allow engineering teams and AI systems to analyze live and historical location information.
The collaboration extends a relationship lasting more than 20 years. What makes this phase noteworthy is its explicit focus on AI agents—not simply analysts working with maps and dashboards.
Giving AI systems spatial context
Generative AI can process enormous amounts of text, but operating in the physical world requires an understanding of where things are, how places connect and how conditions change over time.
A logistics agent cannot recommend a dependable route based only on an address. It may need current traffic, vehicle restrictions, historical travel patterns and an authoritative road network. An infrastructure agent may need to connect asset records with environmental conditions, nearby construction and access routes.
HERE brings frequently updated map and mobility content, while ArcGIS provides technology for managing, analyzing and visualizing spatial information. According to the announcement, the planned capabilities will support reporting, decision-making and agentic workflows across several industries.
Reliability matters more than fluency
The partnership points toward a broader shift: geospatial information is becoming context that software agents can query and potentially act upon.
That creates opportunities in transportation, logistics, public services, insurance and infrastructure management. But access to spatial data alone does not make an AI system trustworthy.
An agent must know when data was collected, what it represents, how accurate it is and whether it is suitable for a particular decision. Live traffic data and historical mobility patterns may be highly useful for route planning, for example, but inappropriate for conclusions they were never designed to support.
This makes provenance, temporal accuracy and human oversight as important as the AI interface itself. Spatially fluent answers can still be wrong if an agent uses outdated maps, combines incompatible datasets or misunderstands geographic scale.
The announcement leaves several practical questions unanswered. The companies have not yet provided a detailed release schedule, pricing structure or complete technical architecture. It is also unclear how organizations will audit agent decisions or control access to sensitive location information.
The significance of the agreement therefore lies less in another promise to “add AI” to GIS. HERE and Esri are positioning authoritative location data and spatial analytics as grounding infrastructure for AI systems.
Whether that produces genuine operational value will depend on something less fashionable but more important: making every automated conclusion traceable to reliable spatial evidence.
