Author: NVIDIA
Translation: PANews
Energy → Chips → Infrastructure → Models → Applications. Every successful application depends on each layer below it, all the way down to the power plant that keeps it running.
AI is one of the most powerful forces shaping the world today. It’s not just a clever application or a single model, but an infrastructure like electricity and the internet.
AI runs on real hardware, real energy, and real economies. It sources raw materials and transforms them at scale into intelligence. Every company will use it, every country will build it.
To understand why AI is unfolding this way, it helps to start from first principles and examine the fundamental changes happening in the computing field.
From Pre-Recorded Software to Real-Time Intelligence
For most of computing history, software was pre-recorded. Humans described an algorithm, and computers executed it. Data had to be carefully structured, stored in tables, and retrieved through precise queries. SQL became indispensable because it made this world operable.
AI breaks this pattern.
This is the first time we have computers capable of understanding unstructured information. They can see images, read text, listen to sounds, understand meaning, and reason about context and intent. Most importantly, they generate intelligence in real time.
Every response is newly created; every answer depends on the context you provide. This isn’t software retrieving stored instructions but software reasoning and generating intelligence on demand.
Because intelligence is generated in real time, the entire underlying compute stack must be reinvented.
AI as Infrastructure
From an industrial perspective, AI can be broken down into a five-layer technology stack.
Energy
The bottom layer is energy. Real-time generated intelligence requires real-time electricity. Every token generated involves electron movement, heat management, and energy conversion into computation. There are no abstractions below this; energy is the first principle of AI infrastructure and the upper limit of how much intelligence the system can produce.
Chips
Above energy are chips. These are processors specifically designed to efficiently convert large-scale energy into computation. AI workloads demand massive parallelism, high-bandwidth memory, and fast interconnects. Advances in chip technology determine how quickly AI can scale and how affordable intelligence becomes.
Infrastructure
Above chips is infrastructure, including land, power supply, cooling, construction, networking, and systems that orchestrate tens of thousands of processors into a single machine. These are AI factories, not designed for storing information but for manufacturing intelligence.
Models
On top of infrastructure are models. AI models understand multiple types of information: language, biology, chemistry, physics, finance, medicine, and the physical world itself. Language models are just one category. Some of the most transformative work is happening in protein AI, chemical AI, physics simulation, robotics, and autonomous systems.
Applications
At the top is applications, where economic value is created. Drug discovery platforms, industrial robots, legal assistants, autonomous vehicles. Autonomous vehicles are AI applications embodied in machines; humanoid robots are AI applications embodied in bodies—the same tech stack, different outcomes.
This is the five-layer cake: Energy → Chips → Infrastructure → Models → Applications.
Every successful application depends on each layer below it, all the way down to the power plant that keeps it running.
We are just beginning this build. Hundreds of billions of dollars have been invested so far, but trillions more in infrastructure remain to be built.
Worldwide, we see chip factories, computer assembly plants, and AI factories being built at unprecedented scale. This is becoming the largest infrastructure project in human history.
The workforce needed to support this build is enormous. AI factories require electricians, plumbers, pipefitters, steelworkers, network technicians, installers, and operators. These are well-paid skilled jobs, and demand exceeds supply. You don’t need a PhD in computer science to participate in this transformation.
Meanwhile, AI is boosting productivity in the knowledge economy. Take radiology as an example: AI now assists in reading scans, but the demand for radiologists continues to grow. This is not a paradox.
The mission of radiologists is patient care; reading scans is just one task in the process. As AI takes on more routine work, radiologists can focus on judgment, communication, and patient care. Hospitals become more efficient, serve more patients, and employ more staff. Productivity creates capacity; capacity drives growth.
What Has Changed in the Past Year?
In the past year, AI has crossed an important threshold: models have become good enough to deliver practical value at scale. Reasoning ability has improved, hallucinations reduced, grounding significantly enhanced. Applications built on AI are now beginning to generate real economic value.
Applications in drug discovery, logistics, customer service, software development, and manufacturing have shown strong product-market fit, creating powerful pull on each underlying layer.
Open-source models play a key role here. Most models worldwide are free; researchers, startups, enterprises, and entire nations rely on open-source models to participate in advanced AI. When open-source models reach cutting-edge levels, they don’t just change software—they activate demand across the entire stack.
DeepSeek-R1 exemplifies this well. By making a powerful reasoning model widely available, it accelerates adoption at the application layer and increases demand for training, infrastructure, chips, and energy below.
What Does This Mean?
Seeing AI as infrastructure clarifies its implications.
AI starts with Transformer LLMs but goes far beyond. It’s an industrial revolution that reshapes how energy is produced and consumed, how factories are built, how work is organized, and how economies grow.
AI factories are being built because intelligence is now generated in real time. Chips are being redesigned because efficiency determines how fast intelligence can scale. Energy becomes central because it sets the ceiling for total intelligence output. Applications are accelerating because their underlying models have finally crossed the threshold where they can deliver practical value at scale.
Each layer reinforces the others.
That’s why this build is so massive, why it touches so many industries, and why it won’t be confined to a single country or sector. Every company will use AI; every country will build it.
We are still early. Most infrastructure doesn’t exist yet, most workforce isn’t trained, and most opportunities are still ahead.
But the direction is clear.
AI is becoming the infrastructure of the modern world. The choices we make now—how fast to build, how broadly to participate, how responsibly to deploy—will shape the face of this era.
