Uncovering the Amazon’s Hidden History: Commencis AI Team Wins 2nd Place in Global Challenge

Commencis AI Team has secured 2nd place in the OpenAI to Z Challenge, a global competition with over 10,000 participants focused on using AI to discover previously unknown archaeological sites in the Amazon rainforest. ​

Stretching across nine countries and covering more than 6 million sq km, the Amazon holds centuries of human history, much of it hidden beneath dense vegetation.

The challenge even caught the attention of National Geographic, which featured the OpenAI to Z Challenge and the groundbreaking AI projects that are shaping the future of archaeology. As noted in the article, “archeological sites, like those in the Amazon, are disappearing faster than we can find them, and AI may help.” The feature highlights how advanced AI models like ours can help researchers “cut through the rainforest not with machetes, but with machine learning,” accelerating the discovery of ancient civilizations hidden beneath the canopy.

Through the GlyphTrack project, Commencis team analyzed LiDAR and satellite imagery through remote sensing, applied machine learning to detect hidden traces beneath vegetation, and used GPT-based reasoning to add archaeological context.​

With this approach, the team successfully identified landscape anomalies beneath dense canopy cover, providing strong leads for archaeologists to explore potential new excavation and research sites in the Amazon region.

Check out the video we put together to give you an inside look at GlyphTrack:

This significant success highlights Commencis’ leadership in AI innovation and demonstrates how technology can support archaeological discovery, preserve cultural heritage, and unlock the hidden history of human civilization beneath the world’s most challenging terrains.

Discover more about GlyphTrack here.