Exploring the Intersection of Beauty Industry and AI

Lets understand why AI models create sterotypes of beauty. 

Artificial Intelligence has transform our way of looking and create images. The beauty industry will change with AI, as others and is not an exception. In order to diversify our AI generated images we need to work more in detail to get out of LLM sterotypes out  of "perfect" beauty and ingeneer a new imaginary of beauty standars.

Here is an example of a AI generated model (Midjourney) of a stereotype of Peruvian woman translated into an advertisment of beauty. We can see the model exagerated the skin texture due to the geolocalization of the subjetc. 

Below the result of Prompt- A close-up image of a woman holding a white tube of cream labeled 'THE CREAM' by Nunzio Saviano. The woman has dark brown, voluminous, wavy hair that cascades over her shoulder, with a glossy, smooth texture. She is wearing a white sleeveless top, and her skin, while appearing healthy and radiant, shows subtle human imperfections: visible pores, fine lines, small fuzz hair, and slight wrinkles, particularly around her eyes and mouth. The skin is softly highlighted by bright, diffused studio lighting, capturing a natural glow. The product tube has a minimalist design with a matte finish, featuring a modern serif 'NS' logo at the top and descriptive text below. The background is a simple, light gray, neutral and clean, allowing the focus to remain on the woman and the product. The overall composition conveys elegance and luxury while embracing realism, with professional lighting enhancing the authentic beauty of her hair and skin. The image is rendered in hyper-realistic, HD, 8K ultra resolution, photographic quality, as if taken with a Nikon camera. 

For other subjects, the prompt can be adapted just adding the context of the subject. Example: 

Subject: An elderly woman from Peru with grey hair, styled in a neat bun. She has deep-set wrinkles, particularly around her eyes and mouth, and age spots on her face. Her skin is tanned, and she has dark brown eyes. The woman is wearing a simple white blouse and is positioned against a neutral grey background. No makeup, with a neutral expression. Lighting: One softbox positioned in front, with two white bounce mirrors on the sides to ensure even lighting. Camera Settings: Capture a headshot from a distance that keeps the model's shoulders within the frame, using a Canon EOS-1D X Mark II with an 85mm lens at f/1.2 and ISO 100. Instructions: The model should maintain a neutral expression. Position her against the neutral grey background with softbox lighting directly in front and white bounce mirrors on the sides. Ensure the face is detailed, but maintain a consistent distance to include shoulders and some space above the head in the shot.

With reference image as follows (AI Generated Image)

From Top Models towards UGC generated by AI.

Even if top model images are still going to use as reference of status and wellbeing, UGC will take its advantages of more candide and natural looks. This is how is possible to be achieved. The AI generated models have classified also all the "natural", "candid" looks to be part of the categories of "real people".  So what is going to be the real beauty standard in a world more than diversified? The answer it will be depending on region and esthetic standards. 

AI Generetaded Image based on Top Model standards for skin care routine:

New trend of User Generated Content will rise in the next years to be a new form of visuals to promote more realible sources of information about the products. 

Keywords for prompting UGC content: Before and After, candid, selfie, Reddit, IPhone...

The beauty standard is not equal in all regions, and not either for LLM and AI models. Not only the AI generated images will answer to this demand but the target will also be part of this transformation. For the following exercise we have choose the following prompt and change the beauty concept: Photo photography,three images side by side，standard three views of a Chinese skincare model 18 years old, hands touching her face.

Option 01

Option 02
Photo photography,three images side by side，standard three views of a (horrible, not good looking, awfull), Chinese skincare model 18 years old, hands touching her face. - Will create this result: 

The result of the exercise make understand that concepts as "model" and "skin care" have highly valules for the positive effect of the LLM allucination. In this case we need to remove those words to get a more candid image but we loose the concept of skin care.

So how we create a high valued keywords a "skin care" to be more natural. Let's try with our first model of exercise 01: "Peruvian woman" - Still not perfect but we gain in authenticity. 

Breaking the standards of beauty.

While diversity and inclusion is one key part of some parts of the industry the new breaking point of AI beauty is to bring any character to any situation of "AI classical beauty standards" Going from this:

To this: 

Conclution:

AI holds the potential to revolutionize the beauty industry, offering unprecedented opportunities to redefine how we perceive, create, and promote beauty. However, this transformation can only succeed if guided by intentional and ethical approaches. The current challenges of inclusivity and bias in AI-generated imagery underscore the need for deliberate efforts to diversify datasets and ensure fair representation of all identities. Without this commitment, AI risks perpetuating harmful stereotypes and excluding underrepresented groups, further entrenching narrow definitions of beauty.

Inclusivity is not just a moral imperative but a strategic advantage in a globalized world where consumers demand representation and diversity. Authenticity is equally critical, as the modern beauty audience values realness and relatability over unattainable perfection. By embracing cultural contexts and celebrating the uniqueness of individuals from different regions, AI can help create a new imaginary that expands the traditional beauty paradigm.

To achieve this, the industry must prioritize transparency in how AI models are trained and used, involve diverse stakeholders in the development process, and remain agile in adapting to evolving cultural standards. The integration of AI should not be about replacing human creativity but about augmenting it, providing tools to envision a future where beauty is inclusive, diverse, and reflective of the richness of human experiences.

Ultimately, AI has the power to democratize beauty, empowering individuals and brands to connect meaningfully across cultural and social boundaries. The journey to achieving this vision will require collaboration, innovation, and a steadfast commitment to ethical practices, but the rewards—a more inclusive and expansive understanding of beauty—are well worth the effort.