As personalization and digital engagement become key drivers of growth in the beauty industry, brands are investing heavily in AI technologies to enhance consumer experiences and streamline product development. The global market for beauty tech is expected to grow at a compound annual growth rate (CAGR) of 19.7%, reaching $8.8 billion by 2027, according to market research firm Research and Markets.
For skin care brands, generative AI tools offer a promising solution to one of the industry’s biggest challenges: substantiating product claims in a scientifically credible and visually compelling way. Consumers increasingly demand proof of efficacy before making purchases, with 74% saying they prefer to see detailed product claims supported by data, according to a 2023 Mintel report.
Recognizing this shift, Haut.AI has launched SkinGPT, a generative AI tool designed to simulate skin care outcomes with scientific accuracy. CosmeticsDesign spoke to Anastasia Georgievskaya, CEO of Haut.AI, about how SkinGPT helps manufacturers and suppliers optimize clinical data, streamline product innovation, and enhance digital consumer experiences.
The evolution of SkinGPT: A new era for virtual skin care try-ons
Since its initial development, SkinGPT has undergone significant advancements. According to Georgievskaya, the platform now offers an innovative solution to simulate the biological effects of skin care products in a way that mirrors real-world outcomes.
“We’re always pushing boundaries to make our technologies faster, more accurate, and more reliable—better in every way for our partners,” Georgievskaya said. “But the big news is the commercial launch of SkinGPT. It’s generative AI-powered technology that brings a true virtual try-on experience to the skin care world.”
While virtual try-on tools are commonplace in makeup, skincare has historically lagged due to the complex, long-term biological changes involved. SkinGPT fills this gap by simulating skin care outcomes over time using clinical data, enabling consumers to “try before they buy” with a higher degree of accuracy.
“SkinGPT can simulate anything that impacts your skin—whether it’s aging, environmental factors, or long-term product use,” Georgievskaya added.
Data-driven precision and simulation accuracy
A key differentiator for SkinGPT is its focus on scientific accuracy. The tool is trained on over 3 million high-resolution images, including clinical trials and real-world selfies from diverse populations.
Haut.AI has also incorporated state-of-the-art generative AI techniques, including generative pre-trained transformers, diffusion models, GANs (Generative Adversarial Networks), and classical computer vision models to ensure the simulations reflect biologically observed trends.
“Our testing processes ensure that every simulation is reliable and validated against real-world outcomes,” Georgievskaya explained. “Unlike traditional visual estimation methods, SkinGPT uses advanced AI techniques to model both biological and visual changes in the skin.”
By comparing simulated results with real clinical trial outcomes, SkinGPT provides brands with greater confidence when substantiating claims, improving consumer trust and credibility.
Synthetic skin data for product innovation
Another key feature of SkinGPT is its ability to generate synthetic skin data, creating hyper-realistic facial images that simulate various skin conditions, such as acne, pigmentation, and wrinkles.
Georgievskaya highlighted the tool’s ability to help brands expand their clinical datasets and refine product formulations before launching full-scale clinical trials.
“For example, SkinGPT can analyze clinical datasets to identify patterns of how skin changes after treatment,” she said. “It then uses these patterns to simulate how a treatment might work on a much larger and more diverse group of people.”
This capability offers brands a cost-effective way to test products on various skin types and demographics, reducing the risk of failure in clinical trials.
E-commerce integration and consumer engagement
SkinGPT is designed to be easily integrated into manufacturers’ and suppliers’ digital platforms, offering new opportunities for personalized consumer engagement.
“One powerful way is by integrating SkinGPT as a virtual skincare try-on tool directly into e-commerce platforms,” said Georgievskaya. “Imagine a consumer shopping for their next serum: they come across a product, click a ‘try it on’ button, upload a photo of their face, and see realistic simulations of how the serum will affect their skin over time.”
Brands can customize the simulations to show results over varying time increments, enhancing the consumer shopping experience. The tool can also work alongside Haut.AI’s AI Skin Analysis platform to deliver fully personalized recommendations and simulations.
Georgievskaya emphasized that SkinGPT can be embedded across multiple touchpoints—from mobile apps and social media platforms to in-store kiosks. The tool’s simulation capabilities also benefit marketing teams by enabling them to create realistic “before and after” images without time-consuming photoshoots or manual editing.
“With SkinGPT, generating visually stunning product effect simulations takes just seconds,” she noted. “We’re confident marketing teams will be saying a big ‘thank you.’”
Regulatory considerations and ethical AI
In a highly regulated industry, compliance with advertising standards is essential when using AI tools like SkinGPT.
“How SkinGPT is used ultimately depends on our partners,” Georgievskaya said. “We provide the tools needed to comply, but brands are responsible for ensuring their final claims align with the specific regulatory frameworks in their regions.”
To help brands navigate these challenges, Haut.AI encourages partners to base simulations on clinical trial data or well-documented research and to clearly label AI-generated images to distinguish them from actual clinical trial results.
Haut.AI also prioritizes data privacy and bias prevention through GDPR-compliant measures and diverse datasets.
“We use enriched datasets and train our neural networks on representative images to minimize any potential bias,” Georgievskaya explained. “This ensures our simulations are accurate and inclusive for everyone.”
Future applications of generative AI in beauty
Looking ahead, Haut.AI sees significant potential for generative AI beyond skin care.
“Hair care try-ons, particularly for hair treatments, is a completely untapped market with massive potential,” Georgievskaya said. “Imagine being able to visualize how a specific shampoo, conditioner, or treatment could impact hair health, shine, or manageability.”
The company is also working on expanding its ingredient simulation capabilities. Georgievskaya shared that brands can now showcase the effects of active ingredients like hyaluronic acid or retinol using SkinGPT, even if they don’t have full product clinical trial data.
“We’re also exploring ways to simulate product combinations—like caffeine and Vitamin K—to deliver even more comprehensive insights,” she said.
Georgievskaya believes SkinGPT offers a competitive advantage for manufacturers and suppliers by making product efficacy more transparent and accessible.
“SkinGPT doesn’t just tell consumers what a product can do; it shows them,” she concluded.
