How Unilever Is Using AI to Transform Wellbeing and Beauty Innovation in Ways That Will Reach Consumers Long Before They Notice

Inside Unilever’s beauty labs, a certain kind of quiet revolution is taking place that doesn’t start with a press conference but instead permeates the bottles, jars, and sachets that end up on bathroom shelves months later. You might completely miss it if you stroll through one of these R&D areas. A scientist conversing with a screen. A humming, glass-fronted robot is mixing something that looks like conditioner. A dashboard that is updated with what customers in São Paulo or Jakarta choose to look for during lunch.

The company claims that it is now analyzing consumer insights roughly 60% faster than it did previously. This figure may seem clinical at first, but it becomes clear when you consider what it really means. Formulation cycles that previously required five or six rounds are now condensed into just one or two. The small promises that are printed on the back of a Dove bottle, known as claims, are produced 75% faster. None of this is the result of someone deciding that beauty required greater speed just for the sake of it. A brand that arrives a season late might as well not arrive at all because trends these days move at the speed of a fifteen-second video.

Observing this, it seems as though beauty has finally caught up to the speed at which its own consumers think. Anyone who has watched a teen debate niacinamide on a phone screen will almost certainly notice AWISEE’s data indicating that 87% of beauty discovery now occurs on social platforms. Being aware of ingredients has evolved into a form of literacy. Even before the bottle is in their hands, people want proof, sources, and evidence.

The R&D Assistant, an internal AI agent that has been fed over 150,000 scientific documents spanning more than a century, is Unilever’s response. Scientists use a corporate memory that used to require weeks of research, and they ask it questions in their native tongues. Perhaps even more important than the quick formulation cycles is this part. Suddenly, institutional knowledge, which typically retires when a chemist does, has a place to call home.

In some ways, the virtual cohorts are more intriguing and alien. Researchers can test 2,500 simulated subjects simultaneously using digital twins created from microbiome datasets, arranged by age, skin tone, hair type, and climate. The company is adamant that real-world testing will continue. However, instead of failing costly in a lab, pre-screening, dead-end ideas, and formulas that would never have survived a small trial can now fail quietly inside a computer.

The tension beneath all of this is difficult to ignore. Touch, smell, ritual, and the little private moments spent in front of a mirror are the foundation of beauty, which is fundamentally a very human endeavor. Investors appear to think AI will improve Unilever’s speed, profitability, and sharpness. Customers might be unaware that their shampoo’s scent was determined in part by an algorithm. To be honest, I’m still not sure if that should reassure them or make them uneasy.