L’Oréal wants to leverage IBM’s generative artificial intelligence (GenAI) technology and expertise to uncover new insights in cosmetic formulation data. Their goal is to facilitate the use of sustainable raw materials, that require less energy and generate less waste.
“To preserve Earth’s natural resources, it is critical to explore renewable, sustainably sourced raw materials when developing consumer products,” said L’Oréal in a press release.
Big data processing
This unique partnership aims to develop a custom AI foundation model [1] engineered to significantly increase the processing capabilities of L’Oréal Research & Innovation teams in every cosmetic category and every region of the world.
The creation of this AI model will use a large number of formulations and component data points to accelerate multiple tasks to be performed by L’Oréal, including the formulation of new products, reformulation of existing cosmetics and optimization for scale-up production.
Trained on a broad set of unlabeled data, foundation models are capable of performing various tasks and applying information from one situation to another. These models have first been applied to the field of natural language processing (NLP) and are now implemented in areas such as chemistry, time series and geospatial modalities.
According to L’Oréal, the formulation foundation model developed with IBM is a first-of-its-kind in the cosmetics industry.
The cosmetics maker expects this tool to help its researchers better understand the behavior of renewable ingredients in cosmetic formulas and thus meet its target of sourcing most of its product formulas based on bio-sourced materials and/or the circular economy by 2030.
“As part of our Digital Transformation Program, this partnership will extend the speed and scale of our innovation and reformulation pipeline, with products always reaching higher standards of inclusivity, sustainability, and personalization,” declares Stéphane Ortiz, Head of Innovation Métiers & Product Development – L’Oréal Research & Innovation.
