Cambridge's Tessera AI Maps Crops From Space to Support Food Security
Tessera, an open-source AI tool that maps crops from satellite images, could offer valuable help to those working to improve food security in the Global South, according to new research from Cambridge.
- Tessera is an open-source AI model developed at Cambridge.
- It is trained on satellite images of the earth and can map crops.
- It can help monitor the impact of climate change on farming.
- New research suggests it could help food security work in the Global South.
- The announcement says such technologies have until now been largely beyond reach.
What Tessera does
Tessera is an AI model developed at the University of Cambridge, according to the Department of Computer Science and Technology. It is trained on images of the earth taken from space. By analysing those images, it can map crops and help monitor the impact of climate change on farming.
The model is open source, so it is not tied to a single provider or a closed commercial service. The announcement does not give technical details such as the model's size, training data or accuracy, so those remain unknown from this source.
Why access matters
The department's announcement says that until now, such technologies have been largely beyond the reach of the people who could use them. The text available for this report breaks off at that point, so the full explanation of who has been left out and why is not covered here.
The framing of the research is clear, though. It suggests that a freely available tool for crop mapping could change who gets to use satellite-based monitoring of agriculture. Mapping which crops are grown, and watching how farming is affected by a changing climate, are central to food security planning.
What it could mean
The researchers present Tessera as a potential aid for organisations and individuals working on food security in the Global South. The claim is cautious: the research "suggests" the tool could help, rather than showing that it already has.
Because Tessera is open source, anyone working on these problems would in principle be able to use the model. The announcement, published on 29 September 2026, does not describe deployments, partners or results from the field. Those details would show how far the tool can deliver on its promise.
Cambridge Computer Science and Technology: How new AI tool could help efforts to improve food security in the Global South
