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

Oracle’s low-code Select AI app makes every visible button part of the production policy

Release 5.0 combines NL2SQL, RAG and agent teams in one APEX interface. Its most consequential control is smaller: administrators can decide who sees exports, deletion, diagnostics and reasoning.

Low-code Oracle AI app with role-based controls over exports, deletion, diagnostics, and reasoning.
AI-generated illustration
By The News Desk· Sep 8, 2026the quick take — two AI hosts go live when you do

Oracle’s Ask Oracle Select AI Release 5.0 is easy to read as a feature expansion: visual agent building, natural-language configuration, an orchestration map, prebuilt agents and no-code profile management for NL2SQL and retrieval-augmented generation. The more consequential production change is the control surface around those features. Oracle now lets administrators govern individual interface actions, including code-editor access, PDF and Excel export, deletion, advanced diagnostics and visibility into agent reasoning.[^1]

A shared interface creates a shared risk boundary

The application brings NL2SQL, RAG and Select AI agent teams into one Oracle APEX experience. Developers can describe an agent in natural language, generate an initial configuration, then refine teams, tasks and tool assignments graphically. An Agent Team Map shows the relationships among those components, while profile management covers model providers, retrieval behavior and validation.[^1]

That consolidation reduces setup friction, but it also means the interface is no longer merely a chat window. It can configure models, expose generated queries, export results and manage agent workflows. Oracle’s button-level controls are therefore not cosmetic personalization; they are the point where a deployment translates policy into the actions a user can actually take.

Start with roles, not defaults

Oracle says administrators can set default conversational styles, NL2SQL and RAG profiles and agent teams. They can also customize branding and navigation for a team or business unit. Those defaults improve usability, but a production rollout should treat them separately from authorization.[^1]

A sensible deployment matrix has at least three roles. Business consumers can ask governed questions and inspect approved outputs without receiving editor, delete or diagnostic controls. Analysts can review generated SQL and export results where policy permits. Builders can edit profiles, tools and agent teams in a controlled environment. Oracle itself gives the example of exposing exploratory features in a demo while hiding delete actions, exports, diagnostics or reasoning details in a production-style internal application.[^1]

Validate the UI as an authorization surface

Before promotion, test every role against every visible action. Verify that a hidden export control cannot be reached through another route, that profile changes require the intended privileges, and that the selected default profile and agent team match the environment. The source describes interface-level controls; it does not report an independent security assessment or measured accuracy gain from Release 5.0.[^1]

The operational lesson is straightforward. Low-code assembly can shorten the path to an agent workflow, but it also moves more consequential behavior into the application layer. Teams adopting Release 5.0 should review the screen the way they would review an API: enumerate actions, assign owners, test permissions and keep development conveniences out of the production role.

[^1]: Oracle, “Ask Oracle Select AI chatbot Release 5.0: From chatbot to low-code AI application platform,” July 30, 2026.

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