NVIDIA CEO Jensen Huang has repeatedly voiced his unwavering belief in the dawn of the Artificial Intelligence (AI) era. He emphasizes that this era is in full swing, propelling a global shift towards NVIDIA computing. This perspective not only underscores NVIDIA's pivotal role in the global AI landscape but also signals the evolution of future technological trends.

NVIDIA's quarterly financial report highlights robust sales of high-performance AI chips, driving significant revenue growth. The data reveals that NVIDIA's revenue for a specific quarter reached $35.082 billion, marking a 94% year-over-year increase. Notably, Data Center revenue alone accounted for $30.8 billion, representing a 112% year-over-year surge. This achievement unequivocally confirms that the AI era is thriving, with NVIDIA, as a leader in AI computing, playing a pivotal role in this transformation.

According to the 2024-2028 Global Artificial Intelligence Chip Industry Development Status and Trend Forecast Report by the China Investment Consulting Corporation, the global AI chip market is experiencing rapid growth and is projected to continue expanding in the coming years. In 2022, the global AI chip market was valued at approximately $44.17 billion, rising to approximately $53.6 billion in 2023 and forecasted to reach $67.1 billion by 2024, with a compound annual growth rate of 15.0%.

Amidst this vast market potential, Huang has been frequently traveling to deliver speeches, revealing his ambitious vision for future AI applications.

01

Huang's AI Empire

Reflecting on NVIDIA's market capitalization growth, it took the company nearly 25 years to surpass the $1 trillion mark. However, it skyrocketed to $2 trillion in less than 9 months and breached the $3 trillion threshold in just about 3 months, even briefly topping the global market capitalization rankings.

This remarkable growth is inextricably linked to Huang's keen technological instincts.

In January 1999, NVIDIA successfully listed on Nasdaq with its innovative graphics processing unit (GPU) technology. Subsequently, NVIDIA introduced its first GeForce series GPU, the GeForce 256 (codenamed NV10), heralding the GPU era for graphics chips.

On December 25, 2000, NVIDIA acquired its main competitor, 3dfx, for $70 million in cash and 1 million shares. This acquisition allowed NVIDIA to assimilate 3dfx's intellectual property and many of its employees, enabling the development of NVIDIA's SLI technology and establishing its dominant position in the graphics chip industry.

As the PC market growth began to stagnate, NVIDIA urgently sought new growth avenues and markets. Building on the success of its GPU products, Huang began leveraging GPU technology for accelerated computing, autonomous driving, robotics, and other domains. In 2006, NVIDIA launched the Compute Unified Device Architecture (CUDA), a parallel computing platform and programming model, providing a general-purpose programming interface for its GPUs. By harnessing the processing power of GPUs, computational performance saw significant improvements. This shift expanded GPUs' applications beyond computer gaming, with widespread adoption in various fields such as molecular dynamics, seismic processing, CT reconstruction, image processing, and, more recently, artificial intelligence.

Furthermore, leveraging its powerful Neural Processing Units (NPUs) and secure frameworks for running AI algorithms, NVIDIA successfully ventured into the robotics market, fostering a multi-billion-dollar industry. With the rise of the cryptocurrency Ethereum, Graphics Processing Units (GPUs) became widely used for cryptocurrency mining. In 2023, NVIDIA initiated the Industrial Metaverse Network Alliance program.

NVIDIA CEO Jensen Huang keenly perceived the potential of GPUs in supercomputing and continuously sought application scenarios that leverage GPU advantages. Following the explosion of the AI chip market, led by ChatGPT, NVIDIA became highly sought-after.

Initially, NVIDIA aimed to create 3D graphics cards for gamers. Its eventual shift towards AI represents one of the most compelling strategic transformations in business history. This transformation not only positioned NVIDIA at the forefront of cultural discourse but also established it as a global leader in AI computing.

02

Huang's Dedication to AI

Having secured a leading position in AI chips, Huang has been traveling extensively this year to deliver speeches, promoting and implementing his AI vision globally.

On July 30, Huang engaged in a dialogue with Mark Zuckerberg at the 2024 SIGGRAPH conference in Denver. They discussed how basic research drives breakthroughs in AI and how generative AI and open-source technology empower developers and creators.

On October 24, Huang participated in a significant dialogue at the NVIDIA AI Summit 2024 in India. NVIDIA's founder and CEO, Huang, had an in-depth exchange with Mukesh Ambani, Asia's richest person and chairman of Reliance Industries. Together, they announced a cooperation plan aimed at strengthening AI infrastructure in India. NVIDIA will collaborate with companies like Reliance Industries, Jio, and Infosys to expand India's computing power 20-fold within a year.

On November 13, Huang engaged in a dialogue with Masayoshi Son, president of SoftBank Group, in Japan. During their conversation, Huang emphasized that Japan must seize the excellent opportunity presented by the latest AI breakthroughs. He believed that no country is better suited to lead the AI robotics revolution than Japan. Huang also revealed that NVIDIA will collaborate with SoftBank to build AI infrastructure in Japan, including Japan's largest AI factory.

On November 23, Huang was awarded an honorary doctorate in engineering from the Hong Kong University of Science and Technology. Following the ceremony, he had an in-depth dialogue with Shen Xiangyang, chairman of the University Council and foreign academician of the National Academy of Engineering of the United States. Huang proposed a compelling viewpoint: the types of robots capable of mass production in the future may be limited to three—automobiles, drones, and humanoid robots. He believes that the Greater Bay Area has unique advantages.

On December 4, NVIDIA signed a partnership agreement with Thai tech company SIAM.AI to help build the country's first 'sovereign cloud.' During his visit to Bangkok, Huang announced this deal and expressed support for AI development in Thailand. Data stored on Thailand's sovereign cloud servers can power local 'AI ecosystems,' such as generative AI chatbots fueled by Thai data.

On December 5, as NVIDIA strengthened its presence in Southeast Asia, Huang agreed to open an AI research and development center in Vietnam. Nguyen Chi Dung, Vietnam's Minister of Planning and Investment, stated at the signing ceremony held in Hanoi on Thursday that this project and the Viettel Group data center utilizing NVIDIA technology would contribute to the development of advanced AI in the country. Huang also announced that NVIDIA would acquire VinBrain, the AI division of Vingroup JSC, marking the 'starting point for a major design center in the future.' The Vietnamese government and NVIDIA signed an agreement to establish the AI research and development center, witnessed by Vietnamese Prime Minister Pham Minh Chinh and NVIDIA CEO and founder Jensen Huang, according to a statement from the Ministry of Planning and Investment.

Huang's recent frequent travels and speeches have undoubtedly sent numerous signals to the tech industry, and his statements are worth contemplating.

03

NVIDIA's Next Step: Humanoid Robots

Heavy industry and manufacturing stand at the forefront of the technological revolution. Investment in robotics is growing at an astonishing rate, signaling the dawn of a new era of industrial intelligence. NVIDIA's continuous innovation in AI is a crucial force driving this change. As a global leader in accelerated computing, NVIDIA is redefining the boundaries of human-computer interaction.

During Huang's Asian tour, he frequently mentioned humanoid robots. In fact, leading global electronics companies are integrating NVIDIA's autonomous robotics technology into their factories, utilizing simulation capabilities in the Omniverse platform to test and validate the new generation of AI in the physical world. This includes over 5 million pre-programmed robots worldwide. More than a dozen leading robotics companies globally, including BYD Electronics, Siemens, Teradyne, and Alphabet's Intrinsic, are integrating NVIDIA Isaac acceleration libraries and other technologies into their software frameworks and robot models to enhance efficiency in factories, warehouses, and distribution centers.

Previously, there were rumors that NVIDIA's humanoid robot project would be implemented in Taiwan. According to industry insiders, NVIDIA's robotics development team recently initiated the first wave of 'auditions' in Taiwan, and several tool machine manufacturers have been approached. This includes inquiries about key component technologies and specifications, as well as interest in joining NVIDIA's robotics supply chain. There are even rumors of establishing an assembly base in Taiwan. NVIDIA's search for and organization of a humanoid robotics supply chain in Taiwan have been confirmed by leading tool machine manufacturers.

In March 2024, NVIDIA unveiled a general foundation model for humanoid robots at the GPU Technology Conference (GTC), aiming to further advance its breakthroughs in robotics and embodied intelligence. Huang referred to this project as the soul of NVIDIA. The humanoid robot project, codenamed GR00T, is a nod to Groot, a superhero from the Marvel Cinematic Universe. In reality, NVIDIA is providing an AI platform for recent humanoid robot manufacturers, including 1X Technologies, Agility Robotics, Apptronik, Boston Dynamics, Figure AI, Fourier Intelligence, Sanctuary AI, Unitree Robotics, and XPeng Robotics, essentially covering all well-known humanoid robot manufacturers.

Additionally, NVIDIA released Jetson Thor, a new humanoid robot computer based on the NVIDIA Thor System-on-Chip (SoC). This SoC includes a next-generation GPU with a transformer engine, utilizing the NVIDIA Blackwell architecture to deliver 800 tera operations per second (TOPS) of 8-bit floating-point AI performance, capable of running multimodal generative AI models like GR00T. Currently, Disney's two small robots, Orange and Green, use Jetson, NVIDIA's first AI chip designed for robots.

It is evident that Huang's next target is humanoid robots.

04

NVIDIA's Strategy in AI Infrastructure and Computing Power Leasing

Beyond robotics, NVIDIA is also positioning itself in AI infrastructure, targeting the computing power leasing market.

The concept of computing power leasing is straightforward; it essentially extends traditional data center businesses. In some ways, it resembles bare metal services in cloud computing, where users are charged for direct access to 'computing power.' For users, this model is ideal for businesses that need to process vast amounts of data but are unwilling to invest heavily in computing power equipment. The most prominent example is AI large model training scenarios, which are currently driving the rise of computing power leasing.

Previously, Punkhash collaborated with NVIDIA to launch a new GPU computing power rental platform, gradually transforming GPU computing power from a 'rich man's game' into a more accessible resource.

Recently, Huang has frequently announced the construction of AI infrastructure in developing countries such as Southeast Asia and India, further unleashing the demand for computing power in these regions. This allows small businesses in these areas to participate in AI competition and benefit from the computing power leasing market.

Moreover, since NVIDIA's most advanced computing power cards are unlikely to be sold to China in the foreseeable future, but NVIDIA does not want to abandon the Chinese market where many companies are accustomed to using its cards. One workaround is to build a computing power center near China and sell computing power services to Chinese companies. This strategy also drives Huang's layout in the Southeast Asian market.

This year, NVIDIA also announced an investment plan worth up to $9 billion, aiming to capture a share of the cloud services market. Previously, NVIDIA's committed investment in cloud computing services was approximately $3.5 billion, and the newly announced investment amount is several times higher.

Long-term, NVIDIA, as a chip supplier, has stood behind the scenes in the cloud computing market. However, its influence is gradually permeating the entire industry chain and may become a variable reshaping the cloud computing landscape. With the rise of GPU cloud services, NVIDIA can not only sell chips but also potentially become one of the largest cloud providers.