Bedrock adds native ServiceNow ingestion with sys ID scoping

AWS announced a native ServiceNow data source connector for Amazon Bedrock Managed Knowledge Base on September 4, 2026. In its announcement, AWS says teams can connect a ServiceNow instance to the managed retrieval-augmented generation (RAG) service without maintaining a custom ingestion pipeline.

Content coverage and availability

The connector crawls knowledge articles and service catalog items, including file attachments. AWS says it handles crawling, metadata extraction, and incremental synchronization after users supply their ServiceNow instance credentials. This provides a managed ingestion path for institutional knowledge already maintained in ServiceNow.

Availability: AWS describes the connector as available now. The page does not specify supported regions, pricing, or whether the release carries a preview or general-availability designation.

Scope the crawl, then check retrieval controls

Teams can restrict ingestion to particular knowledge bases, article categories, or service catalogs through sys ID inclusion lists. For an employee IT assistant, a useful starting point is an explicit list of approved content sources, with attachment review included in the ingestion plan. The inclusion lists are the documented mechanism for narrowing the content entering the knowledge base.

That boundary should not be mistaken for an end-user authorization policy. The announcement does not describe whether or how ServiceNow permissions are preserved during retrieval. Engineers should resolve that question before exposing indexed material to audiences with different access rights.

Incremental sync also does not establish a freshness guarantee: AWS provides no synchronization interval in this announcement. Teams evaluating the connector should check update and deletion behavior against their requirements. The engineering benefit is a vendor-managed crawling path; content selection, access design, and freshness validation remain rollout decisions.