Starts Nov 23Live Cohort 3 of Vibe Coding & AI Agents (Nov 23, 18:30 MSK): 3 weeks, 9 live hands-on sessionsCohort 3 (Nov 23) →
ReleaseInsight · On RadarTNW→ original

Databricks Brings Genie to Slack: Answers About Company Data Right in the Chat

Databricks has made its Genie app for Slack generally available. Anyone can ask a question about company data directly in a work chat, without opening a separate panel. Typing @Genie in a channel or in a direct message brings the answer back in the same thread, with a chart and links to sources such as Confluence and Jira. Each query runs with the asker's permissions, but by default, answers in channels are visible to everyone. Admins should therefore enable private answers before rolling it out.

AI-processed from TNW; edited by Hamidun News
Databricks Brings Genie to Slack: Answers About Company Data Right in the Chat
Source: TNW. Collage: Hamidun News.
Listen to article
Article Fact Sheet & Key Figures
ParameterDetails
Primary sourceTNW ↗
Editor-in-Chief
Key metric #1Genie edits Jira and documents only after the request author confirms
Key metric #2Queries are processed inside the client's Databricks workspace
Key metric #3Databricks warns that answers can be wrong and must be verified

Anyone who has worked at a company with a lot of data knows the ritual: you need a revenue figure or the count of open tickets, and the answer is either buried in a dashboard you need access to, or sitting with an analyst who is booked until Friday. Databricks is trying to remove that barrier by moving its Genie assistant to where employees already spend their day: Slack. The app is now generally available, which means it is no longer a closed test.

How It Works in Practice

Users can message Genie directly or mention @Genie in any channel. The answer arrives in the same thread so it doesn't clutter the main feed. The reply includes links to sources, which can be data from the Databricks platform itself, as well as documents from Google Drive and Confluence, and tasks from Jira.

Beyond text, Genie can build tables and charts, analyze uploaded files, and send reports on a schedule. Picture a product manager who gets a summary of active users in the channel every Monday without contacting an analyst. Through connectors, the assistant can also take action: update a document, leave a comment on a ticket, or change a record in another service. Before making any such change, it asks the person who sent the request for approval, and it changes nothing without that confirmation.

Access Rights and the Main Pitfall

Every query runs with the permissions of the person who asked it. If an employee has no access to a salary table, Genie does not use it in the answer. On paper, this looks reliable.

But there is a catch worth noticing. In channels, answers are public by default. So if a developer with access to financial data asks a question in a shared channel, the figures in the answer will be visible to every channel member, including those who lack such access. The system itself does not break access rules, but human error still does its damage. Administrators can enable a mode in which all answers are private. For companies with sensitive data, this is likely worth doing before a broad rollout, not after the first awkward incident.

As for processing privacy, Databricks states that questions are processed inside each customer's workspace and that the service relaying messages between Slack and Databricks does not store them. This is the company's own claim; the material contains no independent verification.

Why It Matters for the Market

Databricks is moving into someone else's space: Slack belongs to Salesforce, which already has its own Slackbot working in channels. That means assistants from different vendors now compete inside one messenger, and companies will have to decide which of them to trust with access to data and with the right to change work tools.

This fits a broader problem: AI agents are spreading quickly through organizations, and according to data the source cites, most IT directors admit they do not fully control them. An assistant in a messenger that reads data and writes to Jira is exactly such a case.

What to Do in Practice

First, before turning on Genie, check the privacy settings for answers and decide where public answers are acceptable. Second, do not disable the confirmation step for actions in Jira and other systems, even if it seems unnecessary. Third, treat answers as a draft: Databricks itself warns that Genie can make mistakes and asks users to verify results before acting on them.

If your team already keeps data in Databricks, the integration could noticeably cut the number of small requests sent to analysts. But the gain in convenience comes with new risks of internal data leaks and of bad decisions made on the basis of a confident but incorrect answer.

Story

Key Questions Answered

How It Works in Practice?

Users can message Genie directly or mention @Genie in any channel. The answer arrives in the same thread so it doesn't clutter the main feed. The reply includes links to sources, which can be data from the Databricks platform itself, as well as documents from Google Drive and Confluence, and tasks from Jira. Beyond text, Genie can build tables and charts, analyze uploaded files, and send reports on a schedule.

What are the key details on Access Rights and the Main Pitfall?

Every query runs with the permissions of the person who asked it. If an employee has no access to a salary table, Genie does not use it in the answer. On paper, this looks reliable. But there is a catch worth noticing. In channels, answers are public by default. So if a developer with access to financial data asks a question in a shared channel, the figures in the answer will be visible to every channel member, including those who lack such access.

Why It Matters for the Market?

Databricks is moving into someone else's space: Slack belongs to Salesforce, which already has its own Slackbot working in channels. That means assistants from different vendors now compete inside one messenger, and companies will have to decide which of them to trust with access to data and with the right to change work tools. This fits a broader problem: AI agents are spreading quickly through organizations, and according to data the source cites, most IT directors admit they do not fully control them.

ZK
Zhemal Khamidun — CPO at Alpina Digital, product manager of AlpinaGPT. Building AI products since 2019. Hamidun News is a pet project that grew into a production aggregator with 6,000+ readers.
✦ Private AI Builders Community · Vibe Club

Ship AI Faster with Founders, Engineers & Weekly Architecture Reviews

Inside Vibe Club (hamidun.com/club) — weekly live masterminds with Zhemal Khamidun, Tech Radar, full LMS access, and 370+ production Claude Code skills.

Join Vibe Club →Explore Hamidun Academy
What do you think?
Loading comments…