AI Revolutionizes Nazca Geoglyph Discovery

The Nazca geoglyphs in Peru have intrigued people for generations. These massive designs, carved into the desert floor over 2,000 years ago, offer a window into the lives, beliefs, and practices of the ancient Nazca civilization. Despite their allure, uncovering these geoglyphs has always been a slow and challenging process. Now, with the help of artificial intelligence (AI), researchers are changing that—and in the process, transforming how we explore and understand history.

Traditionally, finding new geoglyphs required aerial photography and tedious field surveys. The Nazca region is vast, and many geoglyphs are faint or damaged, making them easy to miss. AI has introduced a game-changing solution. By analyzing high-resolution aerial images, an AI model scanned hundreds of square kilometers of the Nazca desert and identified over 300 previously unknown geoglyphs in just six months. This discovery nearly doubled the number of known figurative geoglyphs and was achieved at a pace previously unimaginable.

The newly found geoglyphs offer exciting insights into Nazca culture. Researchers have identified two distinct styles: line-type and relief-type geoglyphs. Line-type geoglyphs, the ones most people associate with the Nazca Lines, are large—often stretching 90 meters—and feature wild animals like birds and monkeys. These designs are linked to straight paths and geometric shapes, suggesting they were part of community rituals, possibly involving pilgrimages or ceremonies.

In contrast, relief-type geoglyphs are much smaller, averaging about nine meters. These depict humans, domesticated animals like llamas, and symbols such as decapitated heads. Found near ancient trails, these designs seem to have been created for smaller, more personal interactions—perhaps as markers or messages for those traveling along the paths. The difference in style and placement reveals the Nazca people’s ability to balance large-scale community projects with individual or small-group expressions.

The use of AI was key to these discoveries. Unlike most AI models, which are trained on huge datasets, this project started with only a few hundred examples of geoglyphs. Researchers got around this limitation by fine-tuning pre-trained AI systems and using creative techniques to expand their dataset. The AI didn’t just identify geoglyphs; it ranked how likely each one was to be authentic, helping archaeologists focus their efforts on the most promising sites. Combining AI with expert fieldwork proved to be an incredibly efficient way to uncover the past.

These discoveries are about more than just increasing the count of known geoglyphs. They highlight the importance of preserving cultural heritage, especially as climate change and human activity threaten ancient sites like these. By identifying and documenting geoglyphs faster than ever, researchers can help protect them for future generations. Beyond preservation, the findings deepen our understanding of the Nazca people, shedding light on their rituals, creativity, and daily lives.

This breakthrough is a glimpse into what’s possible when technology meets curiosity. AI is revolutionizing archaeology, allowing researchers to uncover hidden histories at a pace and scale that was unthinkable even a decade ago. Whether it’s mapping lost cities or finding artifacts buried for centuries, the potential is enormous. The Nazca geoglyphs remind us that even as we embrace modern technology, there is still so much to learn from the ancient world—and AI is helping us do just that.

Sakai, M., Sakurai, A., Lu, S., Olano, J., Albrecht, C. M., Hamann, H. F., & Freitag, M. (2024). AI-accelerated Nazca survey nearly doubles the number of known figurative geoglyphs and sheds light on their purpose. Proceedings of the National Academy of Sciences, 121(40), e2407652121. https://doi.org/10.1073/pnas.2407652121 

Silverman, H. (1994). The archaeological identification of an ancient Peruvian pilgrimage center. World Archaeology, 26(1), 1–18.  

Sakai, M., Olano, J., & Takahashi, H. (2021). Líneas y cerámica en las Pampas de Nasca, Perú, 2011–2013. Yamagata University Press.