ChatGPT: A New Chapter in AI Transformation

ChatGPT: A New Chapter in AI Transformation

The tech market over the past year has been dominated by AI since ChatGPT was introduced to the world in November 2022. Like the metaverse before it, there have been countless products and articles, and the media has been rushing to introduce them. Many companies have been thinking about how they can apply the new ChatGPT and generative AI to their field work. Some companies, such as Microsoft, are leading new innovations by applying generative AI to their products rather than ChatGPT. It felt like an era of great transformation through AI. Beyond digital transformation, we now talk about AI transformation. In this article, I will introduce the meaning of AI transformation, ChatGPT, and various AI tools.

1. Introduction: A New Chapter in the AI Transformation

Rapid advances in artificial intelligence (AI) technology are revolutionizing many aspects of our lives. At the center of this transformation are conversational AI models with advanced natural language processing capabilities based on large language models (LLMs) such as ChatGPT. They have the ability to generate natural conversations by interacting with users, heralding a fundamental shift in the way humans interact with technology.

The shift to AI is not just a technological advancement; it is also having far-reaching social and economic implications. The emergence of ChatGPT symbolizes this shift, enabling new forms of human-machine interaction that go beyond traditional interfaces and communication methods.

ChatGPT: A New Chapter in AI Transformation
ChatGPT: A New Chapter in AI Transformation

2. ChatGPT’s innovative approach

ChatGPT is one of the major breakthroughs in the field of natural language processing (NLP), with the ability to respond to user questions or commands in a human-like manner. This ability is based on the Generative Pre-trained Transformer (GPT) architecture developed by OpenAI, which is pre-trained on large datasets and then fine-tuned for specific tasks.

When evaluating ChatGPT, the biggest difference between it and similar services before it is that it makes it easy for anyone to use a high-quality conversational service by simply signing up. By building a huge language model to answer any question and opening it up to the public, the service has spread at a rapid pace, reaching 1 million signups in 5 days. It’s also true that the various use cases have gone viral since opening it up to the public.

2.1. Technical foundation

The heart of ChatGPT lies in its Transformer architecture. This architecture plays a key role in understanding context and identifying relationships between sentences. Transformer models use a self-attention mechanism to generate output for a given input. This allows the model to learn how each part of the text interacts with the overall context, which is essential for generating high-quality, natural dialog.

2.2. Scaling applications

ChatGPT has many applications and is constantly expanding. In education, it can contribute to enhancing the learning experience by providing personalized learning and tutoring systems. In business, it is used to improve user experience by automating and improving customer service. It can also assist human efforts and promote creativity in areas ranging from content generation, coding assistance, and creative work.

ChatGPT’s use cases are constantly expanding as AI technology advances. This heralds a future where AI and human interactions will become more natural and efficient, forcing us to reevaluate the impact of AI on our daily lives and professions and explore new possibilities.

These innovative approaches provide deep insights into the future of AI technology and the challenges facing our society. Ongoing research and dialogue is needed to make the most of the positive changes that technologies like ChatGPT will bring, while also addressing the ethical and social issues that may arise.

3. Social impact and outlook

Advances in advanced AI technologies like ChatGPT are expected to have far-reaching impacts across society. While these technologies have incredible potential, they also raise ethical and social questions and require in-depth discussions about how we develop and utilize them.

3.1. Technology and ethics

Advances in AI technology, particularly those related to natural language processing, raise important ethical considerations such as data bias, privacy, and transparency and accountability issues for machine decisions. These are important challenges that must be addressed to maximize the benefits and minimize the negative impacts of technological advances. Data bias can cause AI models to reinforce inaccurate or negative biases about certain groups, making it essential to collect and process data with diversity and inclusivity in mind.

3.2. Jobs of the future

Advances in AI technology have the potential to replace human roles in some specialized fields, which can lead to social problems such as job losses. Recent news stories about people whose jobs have been affected by AI or who have received layoff notices suggest that this problem is beginning to take hold. At the same time, however, the emergence of new technologies and services will spur the creation of new types of jobs. Responding to these changes will emphasize the importance of skills training and retraining, which will play a key role in ensuring that all individuals can benefit from technological advances in the future.

4. Conclusion: The journey to innovation

The emergence of AI technologies like ChatGPT is opening up a new chapter in technological advancement, and we need to work together to responsibly embrace and positively leverage the changes it brings. To realize the full potential of AI technologies and address the ethical and social issues that may arise at the same time, we need collaboration and ongoing dialog among all stakeholders, including researchers, developers, policymakers, and users.

But even apart from these concerns, AI will begin to transform the enterprise. If the first 50 years of the Industrial Revolution was a period of technological change in which machines replaced human manual labor, the next 50 years will be an era in which AI will replace human brain labor. In this article, I will discuss how companies can actually use AI to change their businesses from the perspective of AI transformation, not just digital transformation.

CES 2024 Recap – Key Trends and Tech Innovations

CES 2024 Recap - Key Trends and Tech Innovations

CES 2024 has come to a close, and we’re back with our personal recap of the show as we hit the booths. With so many companies and products on display, it’s hard to cover it all, so we’ve picked 12 key takeaways. We hope you’ll find them useful in your own CES 2024 report.

1. Tech corporatization of traditional companies through digital transformation

Last year, it started with John Deere, and this CES 2024 it was L’Oréal. I think that the steady investment in technology over the past decade, multiple acquisitions, and collaboration with startups have been the driving force behind the explosive response to this keynote. I think that the corporatization of technology by traditional companies such as HD Hyundai and Doosan Bobcat is the need of the hour.

CES 2024 Recap - Key Trends and Tech Innovations
CES 2024 Recap – Key Trends and Tech Innovations

2. AI is everywhere.

There were some GPTs and LLMs as well, but those were driven by interest after ChatGPT appeared, and it seems that the timeframe was too short to actually bring a full-fledged service to CES 2024. This year, some companies showed services or talked about concepts, but I think we still need a little more time to see them at a mass production level, and I think we’ll probably see them next year.

Instead, AI has become a default for most products, and it seems that things that were introduced a few years ago to solve the inherent problems of those products are now commonplace. It’s become a default, and it’s no longer a bragging right? It felt like you had to do it or be left out of the game.

3. Consumer electronics companies choose reality over innovation

There seems to be mixed reviews of global leading consumer electronics companies such as LG Electronics and Samsung Electronics, but based on my experience of attending these companies and overseeing the actual CES exhibition, I think they have made a lot of realistic choices rather than taking on an innovative image. Ten years ago, CES was a place to showcase products that were close to mass production, but just before the coronavirus, it became a race to show more innovation than mass production that year. This part has now returned to its rightful place. Most of the innovative products, such as LG’s transparent OLED T and Samsung’s Bully, have announced plans for mass production this year.

4. Expanding carbon reduction challenges, including new materials and plastic recycling

As the environmental dimension of ESG has emphasized plastic recycling and carbon reduction, we’ve seen more and more companies cite these as their primary business. Large companies are emphasizing these attempts to complete their own closed loop, and Samsung Electronics, LG Electronics, and Panasonic are all examples of this. However, since many startups are research-based, the fact that there are many startups challenging new materials means that we could have predicted changes in the industry or market expansion within a few years.

5. Robots hide behind B2B or special-purpose, home robotics attempts, but domestic conglomerates recycle

After stealing the spotlight last year, robots have taken a backseat this year. However, special-purpose robots such as lawn mowers, snow shovelers, and products that will be sold in B2B such as cleaning fish tanks or sea pollution, and serving were still trying to carve out a market. In terms of actual robots, Chinese companies are the hidden powerhouses, so the lack of a full-fledged CES penetration from them left me feeling a bit lukewarm. In the case of Korean conglomerates, they came out with Samsung Electronics’ Buli, which is a social robot for the home, and LG Electronics’ unnamed but cute robot, which showed advancements over previous robots by understanding conversations and context based on LLM.

However, Samsung Electronics introduced Buli with the same name three years ago, and LG Electronics introduced a rolling robot at CES about five years ago, but did not release it, but this time they announced its release, so it will be interesting to watch the changes in this market. Amazon is also Alexa-based, but they continue to challenge, so it will be interesting to see how usability can change as Alexa expands to LLM technology, and it started as a robot, but I think it will be an interesting point of view to watch as an adjacent area.

6. Expanding the look and feel of a smart home booth

Last year, it is true that emphasizing MATTER gave considerable strength to smart home, but this year, if you look at the exterior only, the exterior has expanded, as if one side of the first floor of Eureka Hall had booths in the smart home category. However, it was quite a bit less because it was an expansion centered on practicality. It could be seen that MATTER is now being organized in the direction of doing built-in, but the part that MATTER did not actually fulfill what it promised.

And the part that wants to protect their territory in the collaboration of large home appliance companies that have not been unified for the past 10 years… Behind it, it is true that there is still a long way to go considering the burden of rising material costs for home appliance companies, but it was still a feeling that a big change continues.

7. The national pavilions expanded to the first and second floors of Eureka Hall, which became a national anthem

To make a long story short, Korean startups occupied half of Eureka Hall, and some countries (such as Germany) that had previously occupied the first floor moved to the second floor. Whether intended or unintended, the startup space gave off a sense of nationalism. I met a lot of people with negative opinions, but personally, I couldn’t deny that it was an opportunity to see many Korean startups at once, and I was able to see domestic startups that have global competitiveness as quantity makes quality.

8. Expanding connections to food tech, beauty tech, and digital health

Foodtech has been around since last year, and beautytech also had a strong showing this year with big names like L’Oreal and Amorepacific. Digital health has always had the largest number of exhibitors at CES, so there was a wide variety of companies this year. If you look at it from the pharmaceutical/bio industry, their level may be perceived as very low, but from a bio perspective, we eat it as food after animal testing, apply it to our skin, and after this accumulation, it leads to pharmaceuticals, so I was looking forward to the connection between them from a tech perspective.

Of course, it is still in its infancy, and it is true that it is humble compared to the companies appearing at the JP Morgan Healthcare conference held at the same time, but even if you narrow it down to digital healthcare, I felt that it is growing compared to before. Of course, few products have been mass-produced or even FDA approved.

9. Continued expansion beyond electric vehicles, e-motorcycles, and e-scooters to all mobility, including electric heavy equipment and electric boats

Despite Hyundai’s new strategy statement on hydrogen, the mobility shift to electricity continued at CES 2024. Caterpillar, for example, was impressive with its heavy equipment being powered by electricity and its energy-related products being offered as separate solutions.

10. Metaverse shrinks, tends toward industrialization

Metaverse is also organizing a separate zone to attract more companies to participate, but we haven’t seen a breakthrough yet. That said, quantity of challenges creates quality, and as in previous years, there was a steady stream of individual solutions, HMDs, haptic aids, and other peripherals – all of which were buried by Apple’s Vision Pro, of course.

Instead, I was interested in the big players, Dassault (France) and Siemens, who are traditional digital twin powerhouses and are steadily expanding their real-world businesses. Dassault’s display of a digital human may not have been very impressive to the average person, but it gave me a glimpse of the technology, data, and sophistication they have accumulated to collect various biometric information and turn it into a twin.

11. Tech for Humanity, Tech for the Underdog

At CES 2024, technologies that solve humanity’s problems and help people with disabilities are my personal favorites, regardless of their marketability, and I’m glad to see them showcased again this year, and that they won so many innovation awards. I saw robotic arms to replace prosthetic limbs, cosmetics and aids for people with hydrocephalus, wearable robots, devices that produce water from the air, and technology that turns seawater into fresh water with less energy, and I wish them all the best.

12. AI such as GPT was expected, but the challenge of AI chips is significant

The talk about AI chips from big companies such as Qualcomm should be approached as a business, but the challenge for AI chips from startups is that the size of this market is growing regardless of success, and I felt that it will expand to On Device, Edge AI, etc. that I expected in the next year or so, and lead to meaningful services based on GPT. Since CES is a hardware-centered exhibition, it is a challenge to put such AI into hardware and tune the performance. I wonder what will happen next year now that we have warmed up this year.

CES 2024 Recap bonus, the return of China

The second largest attendance at CES 2024 was from Chinese companies. This time, the entire Westgate hall above the LVCC North hall was filled with Chinese companies based in China and Hong Kong, and even some of the booths in the back of the hall. The Chinese exhibitors came out with ready-to-sell products, and many of the booths were sales-oriented, and it was great to see the return of the Shenzhen-branded booths.

Of course, there were no Huawei, DJI, etc. representing China, but Roborax, a robot vacuum cleaner powerhouse, was selling well as if it were not a Chinese company.

I’ve been attending CES for more than 10 years and I feel like every time, it’s a similar show, and sometimes I feel like I can’t feel the innovation, but the experience of seeing, touching, and talking to it is really important. I’ll be back next year.

Public cloud: a key component of your digital transformation strategy

Public cloud: a key component of your digital transformation strategy

Public cloud: a key component of your digital transformation strategy. Key Technologies for Digital Transformation: From RPA to Cloud analyzes the speed, agility, cost-effectiveness, and technical advantages of the public cloud, which has emerged as a prerequisite for the fourth wave of digital transformation (DT, DX, digital transformation).

It is no exaggeration to say that cloud is the most fundamental technology for DX. The emergence of cloud services represented by Amazon AWS, Microsoft Azure, and Google GCP can be said to have accelerated the actual DX. It is now a rare sight to see even domestic companies putting servers in IDCs (Internet Data Centers) and visiting them regularly. For companies that aim to provide global services, the emergence of public clouds is a key tool that solves many complex problems at once, like the sweet rain in a drought. How should public clouds be viewed from a DX perspective?

Public cloud: a key component of your digital transformation strategy
Public cloud: a key component of your digital transformation strategy

Speed and agility in the public cloud: Amazon AWS, Microsoft Azrue, Google GCP

One of the biggest selling points of a public cloud environment is speed and agility. In a traditional on-premises environment, building a server starts with ordering hardware. The infrastructure operations staff, in consultation with the development team, assembles the right mix of components for the server’s availability.

This usually takes a few weeks, or even months if the order is placed with an offshore IDC. Once the hardware is received, the next step is to install the operating system and software for the server, connect it to the network, test it for stability, and install it. From there, you use your own server management software to make sure the server is working properly, and if there are any unresolved issues, you travel to the IDC’s site to deal with them. In a public cloud environment, however, this process is much simpler.

You can select the server specifications you need online, and with a few clicks, the hardware is ready to go. What’s more, when you select a server specification, you can sometimes request that the server operating system or some software be installed. In other words, the public cloud is like choosing a ready-to-wear suit, whereas the old way of doing things was like making a custom suit. Because of this immediacy, the public cloud has the advantage of being faster and more agile than traditional methods, which means you can spend less time bringing new services to market and more time testing them. You can also scale to a global service in minutes.

Cost savings and flexibility benefits

Using a public cloud environment can also save you money. Of course, this can be controversial. Recent public cloud environments utilize PaaS (Platform as a Service, a model that provides various IT infrastructures necessary for app development in the form of a platform, and companies only need to worry about app development), which may cost more than a traditional server configuration. However, from the perspective of traditional server operation, cost savings are inevitable until a certain scale is reached. In addition to the pure infrastructure costs of running a single service, there is also a significant cost for operating personnel. With the public cloud, these labor costs can be flexibly managed.

In addition, since the public cloud is not an investment in servers (hardware), you only pay for what you use (pay-as-you-go). This means that for corporate accounting purposes, it is not an “investment” but an “operation” of an asset. In addition, pay-as-you-go allows you to try different business experiments without worrying about the cost if you don’t know what your server usage will be.

In server operations, server specifications are determined based on an agreed transaction between infrastructure personnel and developers, and if the actual usage is less than this, it is an over-investment in the server. In other words, if a server with an availability of 100 is actually operated and only 20 is used, 80 server resources are over-invested.

In fact, these details are rarely reported to the company’s management, and even if they were, they wouldn’t be able to recoup the investment. However, in a public cloud environment, you have the flexibility to handle this situation. If you set up 100 server resources and only use 20, you can downsize to 25 or 30. This scalability and flexibility is also a huge advantage in the opposite situation.

If your service is so successful that you experience a temporary spike in demand for server resources, you can quickly and flexibly scale up your server specifications, and we’ve seen examples where this flexibility has actually reduced server failure rates by more than 50%. Even if you do experience a serious problem with your server resources, your server recovery time will be much shorter than before.

DX Core Technologies and the Public Cloud

There’s another important factor in utilizing the public cloud from a DX perspective. This is that most of the latest technologies for DX, such as artificial intelligence, big data, and the Internet of Things, are implemented in public cloud environments. As we mentioned earlier, it’s important to utilize external resources to build well-crafted technologies instead of trying to build them from scratch. In this context, we also talked about the importance of open innovation. If you want to take advantage of this scalability and fast-moving technology at the right time, you can only do so if your company’s systems are in a public cloud environment.

Whether you’re creating a minimalist product to bring to market or experimenting, the public cloud can give you the most bang for your buck. In the old way, it was difficult to start a new business because you had to acquire a lot of assets, starting with hardware, and the sunk costs were quite high if the business failed, but now you don’t have to worry about that. Of course, how well you utilize the public cloud environment is another matter, and it’s also an internal capability. Nevertheless, the public cloud has become an integral part of the DX landscape.