NVIDIA CEO Jensen Huang: Global Control Over AI Development Pace is Absurd, Supports AI Cloud Computing Industry
NVIDIA CEO Jensen Huang stated at the GTC conference in Berlin on Thursday that the global control over the pace of AI development is absurd and ridiculous, and he supports the AI cloud computing industry, believing that strong demand for computing power will drive companies to address issues in data center construction and operations.
He revealed that he had urged UK AI cloud service provider Nscale to push forward with its projects, stating that the company has quality clients, with the key being to secure sufficient operating funds.
However, the AI infrastructure industry is facing construction delays, financing pressures, and concerns over investment returns.
Nscale is encountering bottlenecks in data center construction and power supply, and its proposed $30 billion IPO valuation is also under scrutiny; another NVIDIA investment target, Firmus Grid, has abandoned its $30 billion IPO plan.
Meanwhile, AI-related stocks have recently come under pressure, with market concerns rising over the high capital expenditures and cash flow capabilities of new cloud computing providers.
To accelerate data center deployment, NVIDIA launched the modular data center product DSX Pod, delivered in container form, at the Berlin GTC conference, aiming to simplify equipment integration and shorten construction cycles, though power supply, site conditions, and supporting facilities remain key constraints for project implementation.
The market focus is no longer just on whether AI computing power demand can continue to grow, but also on whether related companies can convert that demand into actual revenue and cash flow.
Under the criticism of control and product launch events, capital is focusing on modular data centers and cloud service providers with sufficient operating funds, benefiting NVIDIA and DSX Pod, while Nscale and Firmus Grid, facing financing difficulties, are under pressure.
Source: Public Information
ABAB AI Insight
Jensen Huang leads NVIDIA in expanding GPUs from gaming to AI computing power, repeatedly promoting data center product iterations at events like GTC, forming a path driven by hardware acceleration demand, including investments and collaborations with cloud service providers.
This time at the Berlin GTC, he criticized global control over the pace of AI and supported cloud computing while urging Nscale to secure operating funds and launch the DSX Pod modular product, motivated by the need to drive solutions to construction and operational issues, shortening deployment cycles.
Similar cases include NVIDIA's previous public statements on data center bottlenecks and other AI cloud service providers delaying IPOs due to power supply and financing issues; currently, the AI infrastructure is in a control phase transitioning from computing power demand to realizable revenue and cash flow.
This essentially belongs to the restructuring of the industrial chain, shifting data centers from traditional construction to container modular delivery, with the mechanism being that power and site constraints still limit implementation, making operating funds and cash flow critical, shifting pricing power from mere computing power demand to products and service providers that can match construction cycles.
ABAB News · Cognitive Law
Demand drives solutions to construction issues
Modularity shortens deployment cycles
Cash flow tests high capital expenditures