This week, one of the most important technology events in the world took place, offering a sneak peek into the future.

The annual Consumer Electronics Show (CES), held in Las Vegas, Nevada, has been showcasing emerging technologies since its inception in 1967 in New York City.

Over the years, CES has introduced numerous groundbreaking devices. In 1970, the first video cassette recorder (VCR) was unveiled. The Nintendo Entertainment System made its entrance in 1985, followed by the Digital Versatile Disc (DVD) in 1996 and the original Xbox console in 2001.

This year, CES featured over 4,500 exhibitors, including around 1,400 startups. Analysts and investors are particularly focused on announcements from prominent companies during this event.

At the beginning of CES 2025, all eyes were on NVIDIA Corporation (NVDA) as CEO Jensen Huang delivered the opening keynote on Monday evening. Huang captivated the audience by outlining NVIDIA’s strategies to maintain its leadership in artificial intelligence (AI) for the coming years.

In this edition of Market 360, I’ll summarize the key points from Huang’s speech, discuss how these developments could impact AI, and explore potential investment opportunities.

Introducing Project DIGITS

If you’re interested in the full details of Huang’s speech, the entire presentation is available on YouTube. However, here are the highlights.

First up is NVIDIA’s latest Blackwell-based chip, the GB10 superchip. This chip is central to Project DIGITS, a compact, personal AI supercomputer.

Huang revealed that Project DIGITS operates with 200 billion parameter models and achieves one petaflop of AI performance. Essentially, it can run AI models comparable to ChatGPT directly from a desktop setup using a regular power outlet.

Targeted at AI researchers and data scientists, this device will be priced at $3,000. It could democratize access to advanced computing by making high-level AI capabilities available to smaller organizations and researchers who typically rely on costly cloud services.

Shipping for this innovative device is expected to begin in May 2025.

The Innovative Cosmos Platform

Next, Huang introduced a new platform called Cosmos.

This platform aims to assist developers in running simulations for physical AI systems. It employs world foundation models (WFMs) that mimic real-world conditions. NVIDIA states that companies can “build, evaluate, deploy, and simulate physical AI models more swiftly while reducing risks of real-world testing and validation.”

Notably, Cosmos is the first WFM developed, utilizing 20 million hours of video data to understand complex physical dynamics. The first batch of these models is currently available to developers.

While the technical terminology surrounding Cosmos may be complex, its importance is clear: it represents an AI that comprehensively understands real-world mechanisms. This has significant implications for industries such as autonomous vehicles (AVs) and humanoid robots, both set to benefit from its capabilities.

Huang emphasized NVIDIA’s comprehensive offerings to the automotive sector, which include training systems for AI, simulation tools, and synthetic data generation systems.

NVIDIA is already collaborating with most major automotive manufacturers, recently announcing new partnerships with Toyota Motor Corporation (TM) and Aurora Innovation Inc. (AUR), which prompted Aurora’s shares to surge 30% following the news.

Huang even predicted that AVs will emerge as the first multi-trillion-dollar robotics industry.

Similarly, the Cosmos platform can expedite robot learning for complex tasks and environmental navigation. This innovation could significantly lower costs and time involved in testing humanoid robots, which hold potential utility in areas such as caregiving, healthcare, manufacturing, and disaster response.

In summary, Huang described Cosmos as the “ChatGPT moment for robotics.”

Looking Ahead in January

Overall, Huang’s keynote at CES provides several insightful takeaways regarding NVIDIA’s future and the direction of AI technology.

First and foremost, NVIDIA is poised to maintain its stronghold in the AI sector by continuously innovating rather than relying solely on its existing chip technology. The company explores diverse applications and demonstrates how to effectively implement AI.

Moreover, whether discussing supercomputers, autonomous vehicles, or robotics, it is becoming increasingly clear that AI will play a crucial role in bringing these futuristic concepts to fruition.

A lingering challenge, however, remains: how to manage the massive quantities of data these technologies will generate.

As the new political landscape unfolds with Donald Trump’s potential return to office on January 20, industry insiders will be closely monitoring policies that could impact data management solutions.