live wire
▸Snowflake’s Cortex Agent quickstart routes metrics and transcripts with prompt rules—but documents no authorization boundarySnowflake▸Oracle Select AI generates SQL/PGQ for property graphs—but errors if one AI profile mixes graphs with tables or viewsOracle▸Domo’s data team standardizes MCP builds with shared scaffolding; beta toolkits inherit user permissionsDomo▸Databricks proposes five outcome signals for agentic BI—and says dashboard counts and license usage miss decision qualityDatabricks▸One vibe-coded database app needed 56+ human test hours and three repair rounds despite AI Studio’s verification claimsMancas & Mancas / arXiv▸BigQuery AI.IF’s default cost mode trains a local model for most rows—and drops row-error thresholdsGoogle Cloud▸AWS’s multi-tenant data-agent pattern gives the broker Lambda no data permissions; Cognito roles carry Lake Formation scopeAWS Big Data Blog▸Amazon OpenSearch’s optimized engine adds nested agent traces, but AWS reserves it for log analytics—not vector searchAWS▸Two interpretation-based provenance semantics cannot express SQL bag difference, new theorem showsKolaitis / arXiv▸Databricks Feature Views reuse one definition for training and live scoring, but online lookup still returns latest available stateDatabricks▸Databricks Genie’s Slack and Teams previews default channel answers to public, even when viewers lack source-data accessDatabricks▸Snowflake’s two-tool GraphRAG agent beat a seven-tool graph setup 70% to 60% on 22 ontology questionsSnowflake Engineering▸Semantix opens an Apache-2.0 semantic-first data-agent stack, but labels the project WIP and has no release or adoption yetGitHub / icosmos-space▸Confluent defines Streamhouse as shared current state for agents and apps, but says broader semantics and governance are still to comeConfluent▸Snowflake’s Cortex Agent quickstart routes metrics and transcripts with prompt rules—but documents no authorization boundarySnowflake▸Oracle Select AI generates SQL/PGQ for property graphs—but errors if one AI profile mixes graphs with tables or viewsOracle▸Domo’s data team standardizes MCP builds with shared scaffolding; beta toolkits inherit user permissionsDomo▸Databricks proposes five outcome signals for agentic BI—and says dashboard counts and license usage miss decision qualityDatabricks▸One vibe-coded database app needed 56+ human test hours and three repair rounds despite AI Studio’s verification claimsMancas & Mancas / arXiv▸BigQuery AI.IF’s default cost mode trains a local model for most rows—and drops row-error thresholdsGoogle Cloud▸AWS’s multi-tenant data-agent pattern gives the broker Lambda no data permissions; Cognito roles carry Lake Formation scopeAWS Big Data Blog▸Amazon OpenSearch’s optimized engine adds nested agent traces, but AWS reserves it for log analytics—not vector searchAWS▸Two interpretation-based provenance semantics cannot express SQL bag difference, new theorem showsKolaitis / arXiv▸Databricks Feature Views reuse one definition for training and live scoring, but online lookup still returns latest available stateDatabricks▸Databricks Genie’s Slack and Teams previews default channel answers to public, even when viewers lack source-data accessDatabricks▸Snowflake’s two-tool GraphRAG agent beat a seven-tool graph setup 70% to 60% on 22 ontology questionsSnowflake Engineering▸Semantix opens an Apache-2.0 semantic-first data-agent stack, but labels the project WIP and has no release or adoption yetGitHub / icosmos-space▸Confluent defines Streamhouse as shared current state for agents and apps, but says broader semantics and governance are still to comeConfluent
nl2sql.ai
analysisEnterprise agents

Oracle puts a governed dispatcher in front of 25 Fusion agentic applications

Fusion Companion can choose transactions, approvals and specialist agents from ChatGPT or Oracle’s own interface—but Oracle has not given the new front door an availability date.

By The News Desk· Oct 26, 2026the quick take — two AI hosts go live when you do

Oracle is moving enterprise-agent interaction up a layer. Rather than asking employees to choose an application, workflow or specialist agent, the newly announced Fusion Companion is designed to interpret a business objective and dispatch the transactions, approvals, agents and agentic applications needed to complete it.

The launch sits on top of a larger product milestone: Oracle says 25 Fusion Claw-powered Agentic Applications are now available across Fusion ERP, HCM, supply chain and customer experience. The company’s examples include reconciling accounting exceptions, comparing payroll variances, allocating scarce materials, analyzing procurement leakage and resolving service-performance problems. These are not framed as chat-only assistants; Oracle says the applications combine analysis with deterministic computation and execute approved actions inside organizational controls.

The front door is becoming a policy boundary

Fusion Companion is meant to be the routing layer. Oracle says it uses business context, enterprise data and the requesting user’s permissions to determine which transactions, workflows, approvals and agents should run. It can also coordinate authorized work across other cloud applications, data platforms, APIs and third-party agents while leaving execution and accountability in Fusion.

That architecture matters for data teams because natural-language access is no longer only a SQL-generation problem. A request such as investigating a close exception can involve reading ledger data, calculating a variance, selecting a remediation workflow and submitting an action for approval. The dispatcher must preserve identity and policy across every handoff—not just at the first prompt.

Oracle is offering two interaction surfaces. Employees will be able to reach Fusion Companion through supported external AI tools such as ChatGPT and through Fusion Agentic Home, Oracle’s native interface for defining objectives, delegating work and tracking progress. The company says the same permissions, policies and approval controls apply in either route.

What is available—and what is not yet dated

The availability language is uneven. Oracle explicitly says the 25 Claw-powered Agentic Applications are now available. Its detailed Companion announcement describes what organizations “will be able to” do, but does not provide a general-availability date, pricing, supported-region list or a precise catalog of external AI interfaces beyond ChatGPT as an example.

That distinction should shape rollout plans. Customers can evaluate the named domain applications now, but should not assume that the cross-application dispatcher or external-chat entry point is production-ready in every tenancy. Before deploying it, teams will need to verify how user identity propagates from the chat surface, where approval state is recorded, which cross-system actions are supported and how administrators audit the dispatcher’s agent selection.

The strategic direction is clear: Oracle wants the system of record—not the chat client—to remain the execution and accountability layer. The operational contract for that new front door still needs more detail.

sources

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.

    more from the desk