live wire
IBM makes watsonx Orchestrate AgentOps, custom LLM judging and Bedrock-agent discovery generally availableIBMSchemaGate 0.1.45 fixes broken Oracle ADB wallet connections and an OCI stack pinned 28 releases behindSchemaGatePDI’s Amazon Quick procurement agent grounds spend answers in vendor, category and contract contextAWS Business Intelligence BlogBigQuery’s ML.METRICS example returns 0.84 accuracy but 0.30 macro-F1 on the same 100-row classification queryGoogle Cloud BigQuery docsSchemaGate 0.1.44 auto-selects sentence embeddings, lifting bundled-schema retrieval from 90/98 to 93/98SchemaGateSchemaGate 0.1.43 adds read-only SQL execution with per-principal table checks—and documents unauthenticated client assertionsSchemaGateDatabox adds reusable AI Analyst Skills with personal/company scope, auto-matching and marketplace installsDataboxFabric previews an AI builder for data-agent instructions, source guidance and example queriesMicrosoft FabricDatabricks trains data-agent retriever to stop early or spend bounded extra search steps, reporting 5.8-second latencyDatabricksThoughtSpot adds SpotterCode coding agent to its Visual Embed PlaygroundThoughtSpotLongMemEval-S audit: 67–73% of restore-fixable 80k-budget errors came from evicted evidence under three policiesarXivSchemaGate 0.1.42 adds dimension-aware retrieval and fixes complex multi-table SQL promptsSchemaGateSnowflake agent toolsets can silently drop inherited tools when callers lack accessSnowflake DocumentationLooker’s VS Code extension reaches GA with MCP-assisted LookML generation, editing and validationGoogle Cloud Looker release docsIBM makes watsonx Orchestrate AgentOps, custom LLM judging and Bedrock-agent discovery generally availableIBMSchemaGate 0.1.45 fixes broken Oracle ADB wallet connections and an OCI stack pinned 28 releases behindSchemaGatePDI’s Amazon Quick procurement agent grounds spend answers in vendor, category and contract contextAWS Business Intelligence BlogBigQuery’s ML.METRICS example returns 0.84 accuracy but 0.30 macro-F1 on the same 100-row classification queryGoogle Cloud BigQuery docsSchemaGate 0.1.44 auto-selects sentence embeddings, lifting bundled-schema retrieval from 90/98 to 93/98SchemaGateSchemaGate 0.1.43 adds read-only SQL execution with per-principal table checks—and documents unauthenticated client assertionsSchemaGateDatabox adds reusable AI Analyst Skills with personal/company scope, auto-matching and marketplace installsDataboxFabric previews an AI builder for data-agent instructions, source guidance and example queriesMicrosoft FabricDatabricks trains data-agent retriever to stop early or spend bounded extra search steps, reporting 5.8-second latencyDatabricksThoughtSpot adds SpotterCode coding agent to its Visual Embed PlaygroundThoughtSpotLongMemEval-S audit: 67–73% of restore-fixable 80k-budget errors came from evicted evidence under three policiesarXivSchemaGate 0.1.42 adds dimension-aware retrieval and fixes complex multi-table SQL promptsSchemaGateSnowflake agent toolsets can silently drop inherited tools when callers lack accessSnowflake DocumentationLooker’s VS Code extension reaches GA with MCP-assisted LookML generation, editing and validationGoogle Cloud Looker release docs
nl2sql.ai
newsEnterprise BI

IBM’s managed Cognos launch keeps agentic BI inside the reporting control plane

The AWS-hosted service preserves Cognos reporting, permissions and business context while adding natural-language exploration and managed platform operations.

Cognos agents working inside a governed analytics control plane.
AI-generated diagram
By The News Desk· Sep 5, 2026the quick take — two AI hosts go live when you do

IBM has made Cognos Analytics available as a managed service on AWS, pairing its existing reporting and governance stack with BI agents for summarization, report authoring and natural-language data exploration.

The useful distinction is architectural, not conversational. IBM says the agents remain grounded in Cognos content, security permissions and business context. That gives enterprises a way to add natural-language querying without first replacing the reports, data models, distribution workflows and governance rules that already define trusted analytics.

What IBM is actually selling

Cognos Analytics as a Service moves responsibility for infrastructure, maintenance, upgrades and platform lifecycle work to IBM. Customers retain Cognos capabilities including reports, dashboards, visualizations, data modeling, governed analytics and targeted report distribution. The service is built on AWS and can also be procured through AWS Marketplace.

The agent layer supports three practical jobs: summarizing complex reports, helping authors create a first draft of an analysis, and letting business users explore governed data in natural language. IBM also says Cognos can connect those agents to Amazon Bedrock while preserving Cognos permissions and business context.

The deployment test

For an existing Cognos estate, the first evaluation should not be whether the assistant can answer a polished demo question. It should be whether the managed service preserves the organization’s current reporting assets, governance model, permissions and business processes during migration—and whether natural-language answers resolve against those same controls.

A practical pilot can therefore start with a narrow report library and three user roles. Test whether each role can discover, summarize and query only the content it already has permission to use. Then compare the agent’s answer with the governed report and inspect what happens when the user asks for a field or report outside that role’s access.

The service reduces platform operations, but it does not remove analytics governance. IBM’s launch instead makes the BI control plane the boundary around the agent: the model can help users find and explore trusted content, while Cognos continues to decide which content, permissions and context count as trusted.

Filed by The News Desk. Corrections: desk@nl2sql.ai · Our standards →

comments · 0

    Comments are moderated before they appear. Your email is used once to confirm it is you — never shown, never sold. Corrections and questions get an answer from the desk when we have one.