A hyperscaler is a company that runs computing infrastructure on a vast scale, with data centers around the world, and can add capacity quickly as demand grows. The term is most often used for the largest public cloud providers, such as Amazon Web Services, Microsoft Azure and Google Cloud, and sometimes for other very large operators.
They rent computing, storage and networking to customers and offer managed services on top of them, including machine learning platforms and hosted AI models. Their scale lets them negotiate for hardware, power and land, design custom chips and servers and spread costs across many customers.
AI has increased the importance of the group. Training and running large models require enormous amounts of specialized chips, electricity and cooling, and hyperscalers are among the largest buyers of AI hardware and some of the biggest spenders on new data centers.
When you assess an AI product, check which cloud it depends on, what happens if capacity is short or prices rise, and whether the work can be moved to another provider.