In the contemporary space economy, Earth observation is based on a paradox. Satellites have never produced so many images of the planet, but the artificial intelligence that must analyze them still lacks suitable training data. It is on this precise point that the Austrian startup Another Earth is positioning itself, which is developing technology capable of generating entirely artificial satellite images in order to train AI models used in environmental and geospatial analysis.
The company announces a raising of three and a half million euros, intended to accelerate the deployment of its synthetic data platform for Earth observation. Behind this relatively modest funding lies the ambition to build a simulation infrastructure making it possible to train artificial intelligence models over the entire earth’s surface.
From observation to simulation of the planet
Over the past fifteen years, the space ecosystem has profoundly changed. Commercial satellite constellations have given rise to a new geospatial data industry. Companies like ICEYE have demonstrated that it is possible to produce large-scale radar images of the Earth, used in particular for environmental monitoring, natural disaster management or monitoring critical infrastructure.
But the analysis of these images now relies largely on artificial intelligence. However, machine learning models require considerable volumes of annotated data, which are often difficult to obtain. In many parts of the world, satellite images are rare, expensive, or insufficiently labeled to effectively train algorithms.
🚨 SMARTJOBS
- MISTRAL – Account Executive, Enterprise, France – Paris
- ANTHROPIC – Startup Partnerships – France & Southern Europe
- CONTEXT – HR Director – Human Resources Director
- ECOLE POLYTECHNIQUE – Director/Deputy Director of International Relations (F/M)
- CLAROTY — Sales Development Representative
- FRACTTAL — Account Manager (France)
- BRICKSAI — Founding Growth Manager
👉 Find all our offers on the DECODE MEDIA Jobboard
đź“© Are you recruiting and want to strengthen your employer brand? Discover our partner offers
The technology developed by Another Earth consists precisely of filling this gap. The startup combines generative AI and 3D modeling to produce synthetic satellite images accompanied by associated geospatial data, such as land use or elevation models. These datasets can then be used to train automated environmental analysis systems, whether for deforestation detection, agricultural monitoring or infrastructure analysis.
The approach therefore consists less of replacing satellites than of supplementing real data with simulated data.
Generate data where it doesn’t exist
In many artificial intelligence projects applied to the planet, the main difficulty lies not in the algorithm itself but in the availability of training data. High-resolution satellite images can be expensive and their preparation often requires particularly time-consuming manual annotation work.
Synthetic data makes it possible to circumvent this constraint. By generating fully labeled artificial images, teams can train their models on specific scenarios: isolated areas, rare phenomena or extreme events.
The platform developed by Another Earth can produce large volumes of simulated satellite images, accompanied by their automatic labels and segmentations. This approach makes it possible in particular to create datasets for cases that are difficult to observe in reality, such as certain environmental disasters or geographic situations poorly covered by satellites.
New players in the Earth observation economy
The market in which Another Earth is located is in the process of structuring, at the crossroads of three segments: Earth observation, geospatial analysis by AI and the production of synthetic data. On the one hand, companies like Maxar Technologies, Planet Labs and BlackSky exploit satellite constellations to produce images of the planet and geospatial intelligence services. This data then feeds a second level of players specialized in environmental analysis and applications, like Kayrros, which uses satellite imagery and artificial intelligence to monitor energy emissions or industrial infrastructures on a global scale. Finally, a new generation of startups is seeking to resolve one of the sector’s main bottlenecks: the lack of training data for AI models. Platforms like AgileView, for example, are developing simulation engines capable of generating 3D environments to produce artificial aerial and satellite images for machine learning.
In this ecosystem, Another Earth positions itself on this third layer, the generation of synthetic geospatial data, intended to complement or enrich data from real satellites.
Fundraising to accelerate deployment
To support this strategy, the startup announces a fundraising of three and a half million euros. The financing brings together new investors as well as several existing capital partners, including the funds Rockstart, Inovexus, Stamco AG and Wake-Up Capital, with the support of Austrian public bodies such as Austrian Research Promotion Agency and Austria Wirtschaftsservice.
The company plans to use these resources to accelerate the deployment of its Synthetic Data Engine, particularly in regions where environmental data remains scarce, such as certain areas of Latin America or sub-Saharan Africa.
Founded in Vienna, Another Earth develops technology combining generative artificial intelligence and 3D modeling to produce synthetic geospatial datasets intended for training environmental analysis algorithms. The company was created by Maya PindeusCEO and co-founder, and Felix GeremusCTO and co-founder, specialist in 3D technologies and geographic information systems.
