NVIDIA has acknowledged that spending on AI infrastructure now surpasses its personnel costs. Company representatives said that maintaining and developing AI systems requires ever-increasing computing power, which is driving up the overall cost of the ecosystem.
Brian Catanzaro, NVIDIA's vice president of applied deep learning, noted that the compute costs for AI in his team are now significantly higher than employee expenses. This is because modern models demand enormous power for training, operation, and service support.
This trend is not unique to NVIDIA. Other technology companies also report AI infrastructure costs growing faster than personnel costs. Analysts estimate global IT and AI investments are rising and could reach trillions of dollars in the coming years.
Nevertheless, NVIDIA emphasizes that the rising costs are not seen as a negative factor. Instead, they reflect the accelerated adoption of AI across all technology sectors. The company actively uses its own developments in products and services, including technologies like DLSS.
Management maintains that AI does not replace people but enhances their capabilities. They believe the future involves specialists using AI as a tool to improve efficiency, not as a substitute for human work.