In the middle of the Amazon rainforest, where roads don't exist and the nearest town might be days away by boat, scientists have long struggled to keep track of the wildlife around them. Batteries die. Cameras break. And by the time researchers trek back to collect their data, months or even a full year may have passed. Now, a small device built by Microsoft is changing that.
Named SPARROW — short for Solar-Powered Acoustic and Remote Recording Observation Watch — this tool can watch over remote forests for more than a year without any human showing up to maintain it. Powered entirely by solar panels, it uses artificial intelligence to capture images and sounds from the forest, process that data on the spot using a tiny computer inside the device, and send everything back to scientists via satellite or cellular networks.
Juan Lavista Ferres, who leads Microsoft's AI for Good Lab, came up with the idea after talking with conservationists. "We never realized how bad it was until we worked closely with them to see the whole workflow," he said in an interview. Getting a single set of data once meant a dangerous journey into the jungle, leaving equipment for weeks, then coming back — sometimes only to find the batteries had died or the memory card was full.
SPARROW has now been set up in 11 countries across five continents, from the Amazon to forests in Asia and Africa. In some places, the devices have been running continuously for over a year. Lavista Ferres said his team thinks of the project like building equipment for NASA. "You know you have the device, but you cannot go there to recover it," he explained. That means the devices have to be extremely tough — able to handle heat, humidity, rain, and curious animals.
For conservationists, the impact is enormous. Instead of spending months just collecting raw footage and audio, they can now focus on what actually matters: understanding which species are present, how populations are changing, and how to protect habitats. SPARROW's AI can even help identify animals automatically, flagging rare species or unusual activity for researchers to review.
The project represents a shift in how technology can support environmental work in the world's most hard-to-reach places. With autonomous tools like SPARROW, scientists no longer need to choose between safety and good data — they can get both.