How Artificial Intelligence Is Changing the Way Perfumes Are Made
Givaudan's Carto learns a perfumer's aesthetic preferences and pre-screens thousands of formula variants in hours, leaving the human nose to decide between filtered candidates.
By Julia MorettiFragrenza makes several of the alternatives featured in our guides — here’s how we test.
4 min read
In a laboratory in Geneva, a perfumer sits before a screen rather than a smelling strip. The system in front of her — built by Givaudan, the world's largest fragrance and flavour company — has already evaluated several thousand ingredient combinations against a defined brief, ranked them by predicted olfactory coherence, flagged potential skin sensitivity concerns, and suggested three candidate structures the human nose might find worth exploring. The process that once took months now takes hours. The nose still decides. But what it decides between has been shaped by something that does not sleep, does not tire, and does not carry the accumulated biases of a human creative education.
This is not a vision of the future. This is how a significant portion of prestige fragrance is developed today.
The Tools Reshaping the Industry
Givaudan's Carto platform, perhaps the most discussed AI tool in fragrance development, operates as a composition assistant: it learns from a perfumer's decisions, builds a model of their aesthetic preferences, and uses that model to accelerate the iterative refinement process that sits at the heart of fragrance creation. Where a perfumer might previously evaluate fifty variations of a formula over several weeks, Carto can generate and pre-screen thousands, presenting only those most likely to satisfy the brief and the perfumer's established sensibility.
Firmenich, now merged with DSM into dsm-firmenich, operates its own AI-assisted creation platform under the CreateWith framework, applying machine learning to ingredient discovery as well as composition. The system has been credited with identifying ingredient combinations that human perfumers, working within the patterns of their training and experience, would be statistically unlikely to have considered.
The ingredient discovery application may ultimately prove more significant than the composition assistance. The palette of materials available to perfumers is vast and expanding — new aroma chemicals emerge from research laboratories with increasing frequency. No human nose can maintain an active, intuitive relationship with every available material. Machine learning systems, trained on vast datasets of olfactory characterisation, can identify candidate materials for any given olfactory goal with a speed and comprehensiveness that is simply not humanly achievable.
The Philosophical Debate
Within the fragrance industry, the arrival of AI tools has produced a debate that is both legitimate and occasionally overheated. On one side sit those who view AI as an unambiguous advancement: a tool that frees perfumers from tedious iterative work and allows them to focus creative energy on the decisions that genuinely require human judgment and artistic vision. On the other side, a smaller but vocal group argues that the iterative, exploratory quality of traditional fragrance creation — including its inefficiencies and accidents — is precisely where genuine olfactory discovery happens.
Both positions contain truth. The history of fragrance includes accidents that became classics: compositions that went wrong and became extraordinary. Any system that pre-screens against predicted failure also pre-screens against certain kinds of unexpected success. The question of what AI optimises for is never neutral — it encodes the preferences and assumptions of its training data, which means it potentially encodes the aesthetic consensus of what has previously sold well rather than what might be genuinely new.
Fragrances Shaped by Algorithms
Several notable fragrances released in recent years have acknowledged AI's role in their development, and the transparency has been instructive. In each case, the AI contribution was described as structural and screening-focused rather than creative in the authorial sense. The fragrance was conceived by a human. The brief was written by a human. The final evaluation and approval was a human decision. What changed was the speed and comprehensiveness of the middle process.
The smart delivery side of this story is equally significant. The Estée Lauder partnership with Exuud, a smart fragrance delivery technology company, points toward a future in which AI does not merely help create fragrance but helps modulate how it is experienced — adjusting intensity based on environmental factors, time of day, or even biometric data. The fragrance as responsive, adaptive experience rather than fixed object.
What This Means for Quality and Creativity
The honest answer is that it is too early to reach definitive conclusions. AI tools have demonstrably accelerated development timelines and reduced the cost of iteration, which in principle allows more creative risk-taking. Whether that potential is being realised, or whether speed-to-market pressures are instead filling the efficiency gains with volume rather than quality, depends on decisions being made by individual houses and individual perfumers.
What seems certain is that the conversation around authorship in fragrance — already complicated by the role of brief-writers, evaluators, and marketing teams in shaping what a perfumer creates — has become considerably more complex. The nose that signs a fragrance increasingly works in collaboration with systems that are neither human nor neutral. Understanding that is the beginning of a more honest relationship with what we are actually wearing.
Discover at Fragrenza
Whatever the tools used to create them, the fragrances that matter are the ones that work on your skin — and discovering those requires trying before committing. Fragrenza's samples pack is designed precisely for this: a low-risk way to explore multiple fragrances on your own chemistry before investing in a full bottle. As AI continues to shape what the industry produces, the most interesting new releases often sit in less-visited territory — our Latest Products collection keeps you current with what is arriving at Fragrenza. And for those interested in the creative edge of contemporary perfumery, our Niche Fragrances collection represents the segment of the industry where human artistry and technical innovation meet most deliberately.
