The increase in AI adoption is an undeniable future in our world, but its performance is based on the quality of the data and accuracy of the training approach. Where AI is driving change across all areas of our lives, those who are influencing its design or use must act to ensure that we’re creating and adopting responsible AI.
In the world of skin analysis and AI-driven skincare routine recommendations, the ultimate goal is to build AI that can recreate the decisions of a professional ethically and without bias. At Renude, we have worked with an extensive team of dermatologists, aestheticians, formulation chemists and data scientists to build AI that can ‘think like an expert.’ Including the most qualified, ethical and experienced professionals in every aspect of the AI building process has been of paramount importance to us, in order to deliver results.
How do you build AI so it makes expert skincare recommendations?
By collecting hundreds of thousands of real expert skin assessments and routine recommendations.
Building AI starts with a vast amount of data; in our case, images, skincare questionnaires and recommendation data from our experts.
Renude launched its B2C service in 2020 and have collected over 3 million data points since then from which to build our AI from. As this data is proprietary, we have been able to standardise this to give us the highest quality of data as a starting point.
Once enough data is collected, it must be sorted and labelled. Image labelling has been a key part of our development, and involves manually reviewing every image to label which of the 30 different parameters that are present. Our labelling team are doctors and dermatologists who are using guidelines created by Consultant Dermatologist Dr Justine Kluk, also Scientific Advisor to Renude. Working with medical professionals using a unified labelling approach results in maximum accuracy.
What makes Renude unique vs other skincare AI platforms?
Unique to Renude, we’ve been validating our recommendations since we launched our B2C service, renude.co, four years ago. Here, customers continually return to the platform after using their routine, and let us know how they got on. Each piece of feedback drastically improves the intelligence of our AI. It allows us to visually understand how their skin has changed, as well as further our understanding of the customer experience with scent, texture and efficacy. We’ve collected over 60,000 review responses with 94% seeing visible improvements in their skin from their new routine – all based on our AI-driven recommendations.
To ensure we’re at the cutting edge of skincare, we have in-house formulation chemists who have worked to manually build a proprietary database of skincare ingredients, mapping exactly how each one impacts a formulation. Using this knowledge, alongside over 100,000 recommendations that have been collected from our team of licensed aestheticians, we’re able to create complete, step-by-step skincare routines using any set of products.
We’ve ensured that we’re building AI responsibly by working exclusively with qualified skin and skincare professionals alongside our team of data scientists and engineers. This means we can truly say we’ve built AI that can think like an expert. And when consumers feel they are getting a high quality experience, trust increases, leading to stronger conversion, retention and lifetime value – so get in touch to see how Renude AI can work seamlessly for you.
