Wildfires can gain dangerous momentum before anyone knows they’ve started, making rapid detection essential to limiting impact. ALERTCalifornia, a UC San Diego-based statewide wildfire monitoring program, provides earlier visibility through a large network of high-resolution cameras and sensors to monitor California’s landscapes in real time.

Through a research collaboration, the Microsoft AI for Good Lab works with ALERTCalifornia to apply advanced AI and Azure cloud computing to spot early signals, like smoke, giving firefighters and emergency managers clearer, faster intel. That early insight can improve response time and increase the chances of stopping a small ignition before it grows into something far more destructive—often before the first 911 call.

Beyond strengthening wildfire response, the collaboration is also opening new opportunities for long‑term environmental monitoring. ALERTCalifornia archives camera footage across its statewide network, creating a valuable record that researchers can use to better understand how climate change is affecting habitats. Endangered species like California condors are frequently spotted within view, offering a window into sensitive habitats over time.

With faster image transmission and near‑real‑time recordings, emergency managers can study smoke and fire behavior in the earliest moments. When every minute matters, this helps them assess severity and make quicker, more informed decisions.

In Colombia’s Middle Magdalena Valley—one of the most biodiverse regions on Earth—scientists are working to protect ecosystems that are increasingly under threat from deforestation and human activity. Home to thousands of plant and animal species, this landscape holds vital clues to the planet’s health. But for years, the biggest challenge has been sorting through the overwhelming volume of recordings captured by remote cameras and microphones placed in remote locations.

Now, through Project SPARROW—Solar-Powered Acoustic and Remote Recording Observation Watch—an AI-powered solution developed by the Microsoft AI for Good Lab, researchers are turning that flood of information into real-time insight. Solar-powered acoustic monitors and camera traps, equipped with advanced sensors and edge AI models, can detect wildlife activity and potential threats as they happen. Even in the most remote environments, data can be transmitted via low-Earth orbit satellites to the cloud, helping conservation teams protect and preserve nature without needing to physically retrieve devices from the field.

The result is a clearer, more immediate picture of ecosystems that have long remained out of reach. Researchers are already uncovering hidden animal behaviors that offer a new understanding. By combining AI with on-the-ground expertise, Project SPARROW is helping scientists observe the rainforest in entirely new ways.

In Tanzania, giraffe populations have dropped by more than 50% over the past 30 years. To help reverse that trend, the Microsoft AI for Good Lab partnered with the Wild Nature Institute to launch GIRAFFE—Generalized Image-based Re-Identification using AI for Fauna Feature Extraction—an open-source AI tool that uses computer vision to identify individual giraffes. This gives conservationists better data on survival, movement, and reproduction to help stabilize and protect giraffe populations.

In 1956, Dr. Anne Innis Dagg, a 23-year-old Canadian scientist, traveled alone to South Africa to study giraffes in the wild. What she noticed was deceptively simple—no two giraffes share the same spot pattern. Like fingerprints, those markings could be used to recognize individuals over time. Seventy years later, that idea is helping power a new generation of conservation tools.

By turning thousands of field images into usable data in minutes, GIRAFFE gives scientists more time to understand what’s happening in the wild. And its potential goes beyond giraffes—this same approach could be adapted for other patterned species like zebras and whale sharks. GIRAFFE is helping conservationists better understand and ultimately protect the wildlife they care for.