Hyperscaler

The word describes scale, not a particular AI model. A hyperscaler can supply the computing and storage behind an application, whether or not that application uses generative AI. People also use the term for hyperscale data centers themselves; in a vendor discussion, establish whether they mean the provider or the facility.

Example: a manufacturer could rent computing capacity and document storage from a hyperscaler for an internal maintenance assistant, instead of building its own server facility. The assistant’s model, search system, and permissions are still separate choices.

Using a large cloud provider does not transfer every security responsibility to it. AWS’s shared-responsibility documentation, for example, distinguishes protection of cloud infrastructure from customer responsibilities such as data management and access configuration. The division depends on the service selected.

When assessing an AI proposal, ask which provider runs it, where the relevant data will be stored, which services it uses, and who controls access. Those answers are more useful than the label alone.

See AI vocabulary in plain English and the definition of a graphics processing unit.