What to Anticipate from the Largest Event of the Year for Nvidia?

CEO Jensen Huang of the SAP Center in San Jose, California, will give the keynote address at Nvidia’s (NVDA) annual GTC conference on March 18. Through March 21, the event will include training and seminars for engineers and developers, robotics and AI forums and debates, and company booths showcasing real-world applications of Nvidia’s technology.

Huang is anticipated to introduce the Blackwell Ultra, the company’s newest flagship AI processor. The Blackwell Ultra, a more powerful variant of Nvidia’s current Blackwell chip, is expected to outperform the chip from the previous year to accelerate AI model training and operation significantly.

Vera, the replacement for Nvidia’s Grace family of CPUs, and Rubin, the company’s next-generation GPU architecture, should also be covered. The existing Grace Blackwell superchip at the top of Nvidia’s AI processing hierarchy will be replaced by the merged super chip, which will be named Vera Rubin after the American astronomer.

In addition, Huang hinted that he would discuss the follow-up with Rubin on Nvidia’s fourth-quarter results call. Like most chip firms, Nvidia is renowned for giving developers and consumers a comprehensive overview of its roadmap so they can be ready for their following products.

Huang will discuss Nvidia’s most recent software changes about its CUDA platform, simulation technologies, and other topics in addition to the company’s upcoming processors. During last year’s event, Huang highlighted Nvidia’s software initiatives related to humanoid robotics.

Additionally, the CEO is expected to give additional details regarding Nvidia’s entry into the humanoid robots race at a panel the firm is hosting.

Huang will participate in a panel discussion with executives from quantum computing startups Alice & Bob, D-Wave (QBTS), IonQ (IONQ), Rigetti (RGTI), and SEEQC as part of the company’s inaugural Quantum Day, which will take place during GTC.

When Huang stated in January that it would take another 15 to 30 years for practical quantum computers to become a reality, he caused a stir and caused stocks in the field to plummet. Over the past several months, Google (GOOG, GOOGL), Microsoft (MSFT), and Amazon (AMZN) have all revealed their quantum computing processors, which has raised interest in the technology.

According to Google CEO Sundar Pichai, practical quantum computers are five to ten years away. Even experts studying quantum computing disagree on when a quantum computer will be genuinely helpful, making it difficult to predict when future technology will mature. Nonetheless, Huang’s discussion with executives from quantum businesses should be fascinating.

Concerns about possible taxes on semiconductor imports into the US and further export restrictions on GPUs going to China are just two of the many problems faced.

Then came the emergence of DeepSeek’s R-1 AI model, which the Chinese business said was trained on Nvidia processors with lower processing capacity. Wall Street’s increasing concerns about AI expenditure and the return on those significant investments in the technology by businesses like Microsoft, Amazon, Google, and Meta have been exacerbated by this.

As of Thursday, Nvidia’s stock was down 13% for the year. Nevertheless, within the last 12 months, the company’s stock price has increased by 28%.

Huang, however, has dismissed those concerns, claiming that “thinking” AI models, such as DeepSeek’s, respond more effectively to user inquiries when more potent CPUs power them. During his speech, the CEO will discuss his opinions on the subject.