live wire
▸Snowflake agent evals can pin judge versions; unpinned metrics may roll forward and shift scoresSnowflake▸ThoughtSpot makes Spotter 3 default for new instances; it requires download rights and always shares query data with an LLMThoughtSpot documentation▸MotherDuck adds image classification and object location to prompt_jev(), returning typed SQL outputs with confidence scoresMotherDuck▸BigQuery conversational analytics adds Montréal and Toronto; mixed-region agents still default to globalGoogle Cloud Documentation▸Fabric enables Azure OpenAI Copilot access by default, but OpenAI-subprocessor and cross-region processing/storage switches remain offMicrosoft Learn▸Snowflake Cortex Agents now continue by default when named tools are inaccessible, emitting warnings instead of failing the runSnowflake documentation▸Databricks Genie One can use group memberships to tailor answers, while Unity Catalog remains the access-control boundaryDatabricks documentation▸Fabric data agents can query Power BI models with Read alone; Build and workspace access are unnecessaryMicrosoft Learn▸Fabric data-agent Git configs expose column flags, but selecting a table still includes every columnMicrosoft Learn▸Google Data Agent Kit offers remote Spark kernels, but notebook exports can still move results into local Pandas or ArrowGoogle Cloud Documentation▸Cube says production analytics agents need explicit partial-result states, caller-filtered model search and durable artifactsCube▸Thinking Machines names Bridgewater behind ReViSQL-K2.6, citing 92.97% SC-16 accuracy at 15% of frontier-model costThinking Machines Lab▸AWS says a mismatched data/tool/prompt release cut its Field Sales agent pass rate from 78% to 38% in pre-productionAWS Artificial Intelligence Blog▸NL2SQL Multi-Agent ships evaluation artifacts across BIRD, CSpider, Archer and KDD DataSpace—but no software licenseGitHub▸Snowflake agent evals can pin judge versions; unpinned metrics may roll forward and shift scoresSnowflake▸ThoughtSpot makes Spotter 3 default for new instances; it requires download rights and always shares query data with an LLMThoughtSpot documentation▸MotherDuck adds image classification and object location to prompt_jev(), returning typed SQL outputs with confidence scoresMotherDuck▸BigQuery conversational analytics adds Montréal and Toronto; mixed-region agents still default to globalGoogle Cloud Documentation▸Fabric enables Azure OpenAI Copilot access by default, but OpenAI-subprocessor and cross-region processing/storage switches remain offMicrosoft Learn▸Snowflake Cortex Agents now continue by default when named tools are inaccessible, emitting warnings instead of failing the runSnowflake documentation▸Databricks Genie One can use group memberships to tailor answers, while Unity Catalog remains the access-control boundaryDatabricks documentation▸Fabric data agents can query Power BI models with Read alone; Build and workspace access are unnecessaryMicrosoft Learn▸Fabric data-agent Git configs expose column flags, but selecting a table still includes every columnMicrosoft Learn▸Google Data Agent Kit offers remote Spark kernels, but notebook exports can still move results into local Pandas or ArrowGoogle Cloud Documentation▸Cube says production analytics agents need explicit partial-result states, caller-filtered model search and durable artifactsCube▸Thinking Machines names Bridgewater behind ReViSQL-K2.6, citing 92.97% SC-16 accuracy at 15% of frontier-model costThinking Machines Lab▸AWS says a mismatched data/tool/prompt release cut its Field Sales agent pass rate from 78% to 38% in pre-productionAWS Artificial Intelligence Blog▸NL2SQL Multi-Agent ships evaluation artifacts across BIRD, CSpider, Archer and KDD DataSpace—but no software licenseGitHub
nl2sql.ai
guideEnterprise controls

Looker 26.20 turns conversational analytics into three separate admin decisions

The new control split lets enterprises enable Explore conversations, dashboard agents and Python-backed advanced analysis independently—but permissions still determine who can reach each model.

Old single toggle versus three separate Looker admin controls with permission layers.
AI-generated illustration
By The News Desk· Oct 15, 2026the quick take — two AI hosts go live when you do

Looker’s 26.20 rollout gives administrators a more precise way to govern conversational analytics. Instead of one broad switch, Google now exposes separate controls for the Conversations page, dashboard agents and Advanced Analytics. The release is scheduled to roll out from October 12 through October 25, and Google added the control change to its notes on October 14.[1]

Three switches, three different surfaces

The Conversational Analytics (Conversations page) setting controls data agents attached to Explores and direct chats with an Explore. Dashboard agents separately controls default agents on user-defined and LookML dashboards. Advanced Analytics is an additional option that lets data-agent creators enable the feature for conversations.[1][2]

That separation matters because these are not interchangeable risk surfaces. Explore conversations expose governed analytical content through chat; dashboard agents place an assistant alongside curated reporting; Advanced Analytics can translate natural-language questions into Python and execute the generated code to produce analysis and visualizations.[2]

Google’s dependency rules prevent Advanced Analytics from floating free of those data surfaces. At least one of the two Conversational Analytics controls must be enabled before Advanced Analytics can be enabled. Even then, Advanced Analytics is off by default. Existing installations are not simply reset: the Conversations and Dashboard agents controls inherit the prior state of the older combined Conversational Analytics setting.[1][2]

The switches do not replace permissions

The admin page is only the first layer. Users still need gemini_in_looker and access_data permissions for each model they query, plus additional task-specific permissions where applicable. Looker also provides separate Agent Manager and Conversational Analytics User roles; administrators can instead create custom roles that apply those permissions only to selected models.[2]

That creates a useful deployment pattern: platform administrators can enable a conversational surface globally while data owners restrict which models particular groups can query. Advanced Analytics does not introduce a separate elevated role in the documentation; it uses the same permissions required for Conversational Analytics data agents.[2]

For enterprise teams, the practical migration check is straightforward: record the old combined setting before 26.20 reaches the instance, verify the inherited state of both new surface controls, and decide separately whether Python-backed Advanced Analytics should remain disabled. The update improves control-plane granularity, but model-level access still lives in Looker’s role system—not in the new toggles.

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