Light Sensing (2024) and “AI Herbarium” (2025)
Plants_Intelligence. Flower as Antenna Research
AI generated video, live growing plans, timelapse video, 2025
The artworks “AI Herbarium” and “Light Sensing”, complementing the immersive installation “Solarceptors”, present a three-year lupin research through an experimental approach exploring two ‘mechanisms’ of plant adaptation and relation between them — plasticity and evolution — that play key roles in plant adaptation, although they are fundamentally different. Plasticity refers to an organism’s ability to adjust to environmental conditions within a single lifetime via epigenetic changes, while evolution involves long-term, heritable genetic mutations across generations.
“AI Herbarium” is an artistic experiment visualizing evolutionary patterns of flowering plants through AI-generated imagery. It expands the “Plants_Intelligence” research, exploring how plasticity—examined in the “Light Sensing” experiment—relates to evolutionary processes. Diversification, manifesting in variations of shape, size, and color, becomes a crucial survival strategy for flowering plants, potentially aiding adaptation to rapid climate change. As evolutionary scientist Colin Hughes from the Zurich Botanical Garden notes: “I find it fascinating and perhaps reassuring that evolution will drive ongoing diversification of life long after humans have disappeared.”
Based on Hughes’ research on wild lupins in the Andes, where nearly 200 new species have rapidly diversified over two million years, we selected 100 herbarium specimens collected since 1925 from altitudes between 2,500 and 5,000 meters. Sourced from renowned herbaria—including Oxford, Kew Gardens, Harvard, Stanford, the U.S. National Herbarium, ETH Zurich, and the University of Zurich Botanical Garden—these specimens trained AI/ML models to generate images simulating past evolutionary patterns and speculating on future floral diversification. This aesthetic research serves as an open-ended experiment, examining the potential of AI/ML models to create an interface between art and science, and outlining directions for more effective future collaborations and interdisciplinary efforts in researching nature’s response to anthropogenic climate change.
“Light Sensing” focuses on plasticity—examining individual plant behaviors and their intelligent decisions throughout their lifespans. Inspired by plant ecologist Katja Tielbörger’s research at Tübingen University, which demonstrated that plants can plastically respond to light competition through strategies like vertical growth, shade tolerance, or lateral avoidance , we conducted a three-month experiment in the FiBL greenhouse with white lupins. This involved green filters simulating neighbor competition (as in original experiment) and colored light filters to stimulate growth (as my artistic interpretation). The experiment resulted in over 100 3D scans across five growth stages and environmental datasets, forming the basis for a Virtual Reality installation “Solarceptors” exploring flowers as antennas and (light) attractors.
Artists: Rasa Smite and Raitis Smits (2025)
Research: Rasa Smite, in the framework of: Plants_Intelligence. Learning Like a Plant research project (2022–2025), led by Yvonne Volkart (Principal Investigator), funded by the Swiss National Science Foundation, hosted by the Institute for Art, Gender, and Nature, Basel Academy of Art and Design FHNW (http://plants-intelligence.ch).
Scientific expertise:
“Light Sensing” was a collaboration with breeding researchers Christine Arncken, Mariateresa Lazzaro, Monika Messmer at FiBL – the Swiss Organic Farming Research Institute, Switzerland.
“AI Herbarium” was a collaboration with Colin Edward Hughes, University of Zurich | UZH – Institute for Systematic and Evolutionary Botany, Zurich, Switzerland.
AI/ML model training: Jurgis Peters
EXHIBITIONS
Unter Pflanzen, Museum Sinclair-Haus, Bad Hamburg, 2025
