top of page
Writer's pictureDEVIKA MENON 2333126

Beauty Spots in the AI Era: Problems behind Generative Beauty Norms

 

Makeup transcends mere beautification, serving as a dynamic canvas for self-expression.  Throughout history, from the kohl-rimmed eyes of ancient Egyptians to the bold red lips of the flapper era, makeup has consistently mirrored cultural shifts and individual choices.  This historical practice underscores the multifaceted nature of makeup – a tool not just for enhancing physical appearance, but also for conveying ideologies, interests, and even moods.  

 

Traditionally, the dissemination of trends and techniques relied on physical guides. In the early 20th century, beauty companies distributed pamphlets and magazines. Television further revolutionised makeup education with pioneering featuring makeup artist demonstrations. Today, these resources remain valuable for recreating iconic looks.  

 

However, the past decade has witnessed a digital shift. YouTube has democratised makeup education, empowering anyone with a camera to share their techniques, making beauty influencers the new generation of makeup gurus. YouTube has become a platform not just for replicating trends, but also for self-expression, with individuals using makeup tutorials to showcase their unique artistry and personality.  

 

The explosion of beauty content on YouTube is a double-edged sword. On the one hand, you can find a makeup tutorial for any style imaginable, created by people from all over the world. This incredible diversity offers a wealth of inspiration and technique. However, the world of makeup application isn't a one-size-fits-all solution. While YouTube is fantastic for learning, not everyone has the same skin tone, type, or budget as the makeup guru. Recreating a look perfectly might not be possible for you, and figuring out what works best for your unique needs can be tricky. This often leads to two options: shelling out for expensive beauty consultations, or experimenting with readily available commercial products, hoping to find the magic combination through trial and error. There's a clear gap between the endless inspiration YouTube provides and achieving your own perfectly personalised look. This is where AI helps! 


The beauty industry is undergoing a significant transformation fuelled by advancements in technology.  One particularly exciting area is the integration of Artificial Intelligence (AI).  AI is playing a pivotal role in shifting the industry's focus towards a more consumer-centric approach through hyper personalised methods.  By analysing vast datasets and user preferences, AI can anticipate individual beauty concerns and recommend products tailored to each person's unique requirements.  This results in a more personalised experience, with AI-powered tools acting as virtual consultants, suggesting solutions and recommendations suited to individual profiles.   


Leading online retailers are further enhancing this experience by employing cutting-edge AI and Augmented Reality (AR) technology.  This allows consumers to virtually "try on" makeup shades, such as lipstick, before committing to a purchase. Utilising a webcam or smartphone camera with good lighting, AI and AR can simulate how a product would appear on individual features. This innovative approach mirrors the traditional practice of in-store tester applications but offers a significant advantage in terms of sustainability and convenience.   

Similarly, many brands are now offering a more in-depth AI experience. Consumers can complete comprehensive questionnaires detailing their specific beauty concerns.  This data is then analysed by the AI, which generates personalised product recommendations tailored to address their individual needs. While in-person beauty consultations remain popular for tasks like skin tone analysis and colour theory application, they can be a significant financial investment. However, content creators have discovered innovative ways to leverage AI tools for these purposes.  Chatbots, powered by AI like ChatGPT, can analyse users' skin’s undertones and suggest suitable shade matches. Beyond this, creators are also harnessing AI to generate makeup suggestions, explore new styles, and discover novel techniques, further enhancing their creative endeavours. The growing adoption of AI within the beauty creator community is evident. 


The integration of AI within the beauty industry extends beyond virtual tools. Companies now strive to explore the use of robotics to automate specific beauty services, such as manicures. These machines leverage AI technology to identify the customer's nail shape and desired style, after which the machine precisely paints the nails of the customer. Proponents of this approach tout its advantages in terms of speed and cost-efficiency.  Customers would then no longer have to schedule appointments weeks in advance and could have salon-quality manicures on their fingertips. 


While advancements in technology and Artificial Intelligence (AI) offer a plethora of promising applications within the beauty industry, their integration also necessitates a critical discussion regarding potential biases. Historically, the beauty industry has been characterised by the promotion of narrow and unrealistic beauty standards, with rigid gender norms dictating makeup use. However, growing consumer awareness has fostered a positive shift towards inclusivity, with brands acknowledging the diverse spectrum of skin tones, features, ethnicities, and gender identities. This progress is reflected in the wider availability of beauty products catering to these varied needs and a more representative portrayal in marketing materials. Makeup has transcended its role in adhering to established gender norms, evolving into a powerful medium of self-expression for all. 


However, the potential benefits of AI-powered beauty tools could be compromised by inherent biases within the underlying algorithms.  To investigate this concern, a series of prompts were submitted to a popular Generative AI platform - Pixlr. These prompts encompassed a variety of scenarios, including comparisons of “a person with and without makeup”, “smokey eye look”, and “a man with makeup”. The resulting images exhibited a concerning lack of diversity, both in terms of racial and gender representation and the portrayal of “the perfect skin”.  

 



 





 

This outcome underscores the critical issue of data bias within AI tools, particularly regarding the persistence of outdated gender norms surrounding makeup use.  The prompt requesting an image of "a man with makeup" resulted in unrepresentative outputs because the AI was trained on data that did not adequately reflect the growing acceptance of makeup as a gender-fluid form of self-expression. The utilisation of training data that lacks sufficient diversity can lead to the development of biassed algorithms. These algorithms may perpetuate unrealistic beauty standards and fail to provide recommendations that cater to a broader demography.  It is therefore imperative for developers of AI tools within the industry to prioritise the creation of inclusive and representative data that reflect the evolving landscape of the society. 


Conclusion 

While AI is bound to create an impact on each sector, it is necessary to ensure that the intensity and the manner of the impact is only in a controlled manner. Although we know that in the soft AI era that we are a part of, AI cannot replace the human touch in any industry, including the beauty industry. However, since AI is fed on readily available data developed by humans and its inability to think beyond what is fed, makes it imperative for the technology to be developed with broader data and understandings of the human society

10 views0 comments

Comments


bottom of page