[Amorepacific Future Creation Seminar] The Future of AI - Seizing New Opportunities

Abstract

These days we cannot spend a day without hearing about artificial intelligence (AI) or machine learning (ML). Even if we survive a day without hearing about them, every moment we take advantage of AI and ML and we are being taken advantage of by them (whether or not we realize it). In short, they are everywhere.

In this talk, instead of directly discussing AI and ML and their applications, we will start addressing Digital Transformation based on a book written by Thomas M. Siebel. The confluence of four major technological forces – cloud computing, big data, AI, and the internet of things (IoT) – is causing a mass extinction event in industry after industry, leaving in its wake a growing number of organizations that have either ceased to exist or have become irrelevant. At the same time, new species of organizations are rapidly emerging, with a different kind of DNA born of this new digital age, such as, Amazon, Google, Netflix, and Spotify. Successful digital transformations require the mandate and leadership of an organization’s top executives: Digital transformation must be driven from the top down. We will explain why Digital Transformation follows the pattern of Punctuated Equilibrium, i.e., mass extinction and subsequent speciation, not Darwinian Evolution, which is continuous evolution by survival of the fittest. We will provide advice on how CEOs and other senior leaders should prepare for these changes.

We will briefly show the history of AI and how and why it has become the game changer for many industries ever since the dramatic image-recognition milestone of the AlexNet designed by Prof Hinton’s student Alex Krizhevsky for the ImageNet challenge in 2012. Since then, many statistical learning problems that had been long believed to be impossible to solve have been solved by many artificial intelligent (AI) and machine learning (ML) technologies even outside of computer vision field. These fields include (but not limited to) natural language processing (NLP), speech recognition, semantic segmentation, recommender system, anomaly detection, and various time series methods.

The latter part of this talk will discuss how Amore Pacific can continue to be successful and dominate the market by transforming itself into a data-driven company by leveraging the AI/ML technology and its full potential. Based on the discussion I had with people from Amore Pacific in the prep meeting on 21-Sep-2020, we will try to answer questions such as how it should collect, store, analyze, and optimize data, how it can take most out of the data related to image processing (IP), computer vision (CV), manufacturing, and supply chain, and how it can use AI for product personalization.

Lastly, we will discuss one of the most interesting topics in AI: The Singularity. When will the singularity come? Or more importantly, will the singularity ever come (with the architecture of the current statistical learning methods, e.g., deep neural network)? The singularity is a hypothetical point in time at which technological growth becomes uncontrollable and irreversible, resulting in unforeseeable changes to human civilization. I will share my point of view on this scary and intriguing topic, which I hope will encourage dynamic and active discussion among the audience.

Claude’s abstract

This presentation explores the transformative potential of artificial intelligence and machine learning technologies within the context of digital transformation, using evolutionary biology’s concept of punctuated equilibrium as a framework for understanding technological disruption. Drawing parallels between mass extinction events in natural history and corporate disruption in the 21st century, the analysis demonstrates how the convergence of cloud computing, big data, IoT, and AI creates unprecedented opportunities for industry transformation. The discussion emphasizes that unlike previous technological advances driven by IT departments, current digital transformation requires CEO-level leadership and strategic vision to navigate the rapidly evolving landscape where 52% of Fortune 500 companies have been acquired, merged, or bankrupted since 2000.

The presentation specifically examines AI applications in the beauty industry, highlighting how machine learning and computer vision technologies can revolutionize product development, customer experience, and market positioning for companies like Amorepacific. Key applications include virtual product testing through facial analysis, personalized skincare formulation using data analytics, and AI-driven beauty recommendations that provide statistical foundations for attractiveness and product compatibility. Examples from industry leaders such as Sephora’s ColorIQ app, Beauty.ai’s deep learning beauty analysis, and Function of Beauty’s customized hair care demonstrate the practical implementation of these technologies to reduce product waste and enhance customer satisfaction.

The technical foundation covers machine learning perspectives from statistical, algorithmic, and engineering viewpoints, emphasizing the importance of proper ML pipeline implementation including data collection, preprocessing, model training, and cloud-based deployment using platforms like AWS. The presentation concludes by addressing the philosophical implications of technological singularity while maintaining a pragmatic focus on immediate business applications. Through case studies and practical examples, this analysis provides a roadmap for beauty industry leaders to leverage AI technologies for competitive advantage while building data-driven organizational capabilities that can adapt to the accelerating pace of technological change.