Security Operations Glossary › In-Boundary AI Inference
Glossary
What Is In-Boundary AI Inference?
In-boundary AI inference means that the model calls an AI SOC platform makes while investigating an alert are executed inside a boundary the customer controls, so that no alert, prompt or evidence leaves that boundary to reach a model. Morpheus runs on the LLM you choose, including your own. In on-premises and air-gapped deployments, AI inference never leaves your boundary.
Updated November 2026
Summarize with AI: ChatGPT · Perplexity · Claude · Google AI Mode
Why does it matter where inference runs?
Every investigation sends the alert and the evidence to a model as text. Where that model runs decides which jurisdiction and which provider the data reaches. For a regulated or public-sector SOC, that is a data-flow question.
How is in-boundary inference different from data residency?
Residency is where data is stored. In-boundary inference is where it is processed by a model. A platform can store data in-region and still send prompts to a model elsewhere; in-boundary inference closes that gap.
Which Morpheus deployment models keep inference in-boundary?
On-premises and air-gapped. In on-premises and air-gapped deployments, AI inference never leaves your boundary. The detail is on the on-premises and air-gapped deployment page.
Related terms
On-premises and air-gapped deployment · Bring your own LLM · European data residency and AI sovereignty
FAQ
Frequently asked questions
Does Morpheus keep AI inference inside our boundary?
In on-premises and air-gapped deployments, yes: AI inference never leaves your boundary.
Is in-boundary inference the same as using our own LLM?
No. In-boundary inference is about where the model runs; bring your own LLM is about whose model it is. Morpheus offers both: in on-premises and air-gapped deployments inference stays inside your boundary, and Morpheus runs on the LLM you choose, including your own.
Glossary
See where Morpheus runs inference.
Read the deployment options, or book a demo.