Skip to main content

COMPEL Glossary / multi-tenant-ai

Multi-tenant AI

A design pattern that serves multiple tenants from a shared AI system with isolation at the retrieval (namespace), rate-limit, policy, cost-attribution, and evaluation layers.

What this means in practice

Multi-tenant AI requires explicit isolation at every plane of the reference architecture; shared inference and shared embeddings do not excuse shared data visibility.

Synonyms

multi-tenant AI architecture , tenant-isolated AI

See also

  • Data residency (AI) — The requirement that training data, retrieval data, and inference itself occur within a specified jurisdiction.
  • AI reference architecture — A canonical layered model — client, orchestration, model, knowledge, observability planes — that every AI system maps onto.
  • Knowledge plane — The layer of an AI system that stores and serves non-parametric knowledge to the model — through retrieval over vector stores, traditional indexes, and tool-based data access.