Nvidia Shares Fall Despite Latest AI Chip Announcement
Nvidia’s stock price experienced a decrease, despite the much-anticipated unveiling of its next-generation artificial intelligence (AI) chips at the annual GPU Technology Conference (GTC).
On Tuesday, Nvidia presented its new AI chips, intended to power future advancements in robotics and self-driving cars. However, the company’s stock fell over 3% following the announcement. This decline reflects a general risk-off sentiment in the market.
Other stocks in the ‘Magnificent Seven’ group also closed lower during the session.
This year, the AI company’s stock has fallen by 15%. This is influenced by global cautious sentiment and the emergence of Chinese firm DeepSeek’s more affordable AI model. Moreover, the company’s recent quarterly earnings report failed to win investors over. Sales growth showed signs of slowing.
According to Josh Gilbert, a market analyst at eToro Australia, “Investors may see that as an opportunity, particularly with its valuation remaining attractive on the backdrop of ongoing growth.”
The GTC conference, a crucial event for Nvidia, brings together AI developers and investors from around the world. During this week-long conference, the chipmaker needs to convince hyperscalers to continue their significant investment in its next-generation chips.
One key update of Nvidia’s chip line is the successor to its Blackwell supercomputing chip, named Blackwell Ultra. This chip is scheduled to begin shipping in the second half of 2025. The Blackwell series chips, integral to Nvidia’s future growth, began shipping in the final quarter of 2024.
The upgraded version can handle more tasks in the same amount of time as its predecessor. This allows cloud providers to generate 50 times the revenue compared to the prior generation Hopper GPUs.
CEO Jensen Huang stated, “We designed Blackwell Ultra for this moment — it’s a single versatile platform that can easily and efficiently do pretraining, post-training, and reasoning AI inference.”
Another significant announcement was the Vera Rubin system. This next-generation system combines CPU and GPU capabilities and is scheduled to launch in the second half of 2026. Named after the astronomer who contributed to the discovery of dark matter, Vera Rubin is intended as a custom-built supercomputing system. It will manage 50 petaflops while performing inference, over double that of the current Blackwell chips.
Nvidia also announced collaborations with Walt Disney and Google DeepMind on the Isaac GR00T N1 project. This project will accelerate robotics development. Furthermore, Nvidia has partnered with General Motors for future automotive AI and will also collaborate with T-Mobile US and Cisco Systems to build AI-powered 6G network hardware.
Nvidia’s top customers are major cloud providers. These include Amazon, Microsoft, Alphabet, and Oracle Corp. Huang stated that these companies have collectively purchased 3.6 million Blackwell AI chips in 2025. Bloomberg reported that hyperscalers, including Microsoft, Amazon, and Meta Platforms, are projected to spend $371 billion (€340 billion) on AI infrastructure in 2025. The investment is expected to increase to $525 billion (€480 billion) by 2032.
Future AI expenditure is expected to emphasize processing power and inference capabilities. This is inspired by innovations from DeepSeek, rather than developing completely new AI models.
Huang added, “In the last 2 to 3 years, a major breakthrough happened, a fundamental advance in artificial intelligence happened. We call it agentic AI, It can reason about how to answer or how to solve a problem.”
