Hardware, Software, and Models
A model is a component. It still needs a computer.
Hardware, software, and models (Simple Explanation)
You now have the parts of the machine.
Hardware is the physical part. NIST: the physical components of an information system.
Software is the programs, plus the documents that go with them. NIST: computer programs and associated documentation.
An AI model is not a third kind of computer. NIST defines an artificial intelligence model as a component of an AI system. That component uses computational, statistical, or machine-learning techniques to produce outputs from inputs.
A model runs on a computer. It does not replace the computer. It still needs hardware. It still needs software to load it, feed it inputs, and take its outputs.
When someone says "the AI did this," name the pieces. Name the computer. Name the software. Name the model. Name the inputs. Name the outputs. The model is one component.
This is the last hardware-first page. The next page is What is AI?. AI is an umbrella category for machine-based systems that produce outputs from inputs. You can read that sentence with a computer in mind, not as a slogan.
Key Ideas
Hardware: physical components. Source: NIST.
Software: programs and associated documentation. Source: NIST.
AI model: a component that uses computational, statistical, or machine-learning techniques to produce outputs from inputs. Source: NIST.
The model does not replace the computer. It runs on one.
Next question
If a model is only a component, what is the category called AI?
References
NIST (2023). hardware - Glossary. NIST Computer Security Resource Center.
NIST (2023). software - Glossary. NIST Computer Security Resource Center.
NIST (2023). artificial intelligence model - Glossary. NIST Computer Security Resource Center.
Citation Note: All referenced papers are open access. We encourage readers to explore the original research for deeper understanding. If you notice any citation errors, please let us know.