Making scents: could AI help perfumers take pressure off endangered plants?

by | Oct 14, 2026 | Science

Making scents: could AI help perfumers take pressure off endangered plants?

The fragrance industry is exploring artificial intelligence and biotechnology as potential solutions to reduce reliance on harvesting endangered plants for scent compounds. Agarwood, derived from damaged Aquilaria trees, exemplifies the conservation challenge—wild populations have been devastated by illegal logging to meet global demand for the expensive ingredient used in luxury perfumes.

Companies and research institutions are developing novel approaches to address this issue. Osmo, a Google Brain spin-off, has demonstrated the ability to recreate a rare tree’s scent from a small bark sample by analyzing it with a spectrometer and using AI to reconstruct the chemical formula. BioHarvest Sciences has created a stable cell culture from an endangered southeast Asian plant, enabling production of its fragrance compounds without harvesting the original species. The company has announced plans for limited production potentially beginning next year, with production agreements targeting 20 tonnes of fragrance composition.

However, experts caution against overstating the environmental benefits of these technological advances. Some conservationists and natural perfumers argue that creating synthetic alternatives does not necessarily eliminate demand for wild-sourced materials, which remain culturally and commercially prized. Rhian Smith of the Royal Botanic Gardens, Kew, noted that synthetic oudh alternatives would not completely reduce pressure on wild Aquilaria populations. Natural perfumer Mandy Aftel characterized such claims as greenwashing, arguing that the technology does not address underlying causes of species decline.

The scientific foundations for these efforts build on existing knowledge. Synthetic perfumes have been used industrially for over a century, and machines have long identified molecular composition. What distinguishes recent work is the predictive capability of AI trained on millions of human responses to scents. Additionally, researchers have reconstructed scent-producing genes from preserved flowers and collected microbes from clouds as starting points for fragrance development.

Limitations remain significant. Floral scents can contain hundreds of molecules that vary based on plant part, growing conditions, and climate. Researchers acknowledge that recreating scents captures only fragments of complex natural processes involving living organisms and their ecological contexts. Osmo’s founder acknowledged that conservation was not an explicit priority for the company’s work, highlighting the gap between technological capability and environmental motivation.

Article Attribution | Read More at Article Source

Article summary produced by Claude AI