Route AI requests through one secure, governed gateway
Also listed as: Secure centralized AI gateway (Azure & local)
What is it?
Xelence 10.0 will send AI requests through one governed entry point, a single API every AI feature calls. It connects to AI models hosted in Azure or run on your own servers. It finds out what each approved model can do in each environment, and can fall back to a backup model if one is unavailable. Access keys stay in a separate protected store, and each call is designed to get the same policy checks and audit record.
Why does it matter?
When each feature connects to AI services on its own, access keys, settings and data handling end up scattered, which makes AI use harder to secure, audit and change.
One place to control which AI models are used and how they are reached, fewer interruptions when a model is unavailable, and the option to keep AI processing inside your own network where your policies require it.
Generative AI Services and Xelence AI Engine work together. Building AI and running it explains which does what.
Getting it into your solution
- Needs an upgrade to the Xelence release
- Needs configuration
Where people stay in control
Administrators decide which AI models and providers each environment may reach, including any backup model. Every call through the gateway is designed to be recorded, so your team can review what ran and which model answered.
Start thinking about
- Do your data policies require AI processing to stay on your own servers, or is a cloud AI service in Azure acceptable for some uses?
- Who in your organization approves which AI models, including backup models, may be used in each environment?
- Which of your systems would call AI features first?
Terms used on this card
- AI gateway
- One governed entry point that every AI feature calls, so access to AI models is controlled, checked and recorded in one place.
- API
- A defined way for two software systems to exchange data or request actions from each other.
- Connector
- A ready-made link between Xelence and another system or service, such as Salesforce, DocuSign or an identity provider, so data flows without custom code.
- Environment
- A separate copy of your solution used for one stage of work, such as development, testing or production. Production is the live system that staff and constituents use.
- Large language model (LLM)
- The kind of AI model that understands and writes text, used behind AI assistants and agents.