Detecting credential-driven attacks on Databricks
Summary
Databricks system tables and the Detection App's Genie agent can identify credential-driven attacks by spotting behavior shifts across six stages of an intrusion. Use ready-to-run detection and response queries alongside ingress and least-privilege controls to quickly surface compromised accounts and limit what they can reach.
Summary generated by brickster.ai. For the full article, follow the source link above.
Topics
More from Databricks Blog
Air gapped, not data dark: how Bright Machines gives customers live factory dashboards without opening the plant
Bright Machines replaced a fragile nine-service pipeline by using Databricks Unity Catalog federation with AWS Glue to query edge-captured Apache Iceberg data from air-gapped plants in place. This unified architecture makes factory PostgreSQL changes queryable within two minutes, powering 10-minute AI/BI Dashboard refreshes and zero-copy sharing under a single governed platform.
Brickster Voices: Meet Alasdair Anderson
Enterprises are modernizing legacy data stacks to eliminate data silos, scale AI solutions, and streamline complex governance and risk modeling. Get an inside look at how the Databricks Field Engineering team guides this work through a typical week with Lead Solutions Architect Alasdair Anderson.
Databricks Doubles Down on AI Products, Opens New 235,000-Square-Foot San Francisco Headquarters
Databricks has opened a new 235,000-square-foot global headquarters at One Sansome in San Francisco, marking a $1 billion-plus investment to expand its AI research and product development. The new facility houses the research and engineering teams building data and AI capabilities for advanced customers such as Anthropic, Cognition, Cursor, Decagon, and OpenAI.
The new Genie: autonomous AI for complex enterprise work
Genie is now an autonomous AI coworker capable of executing long-running enterprise tasks within your defined guardrails. Supported by Genie Ontology and Genie Search across the lakehouse and business systems, the updated platform relies on Unity Gateway so admins can govern tool access, trace actions, and set spending limits.
Genie with Genie Ontology Delivers 1.6× the Accuracy with Less than Half the Tokens
Genie achieved a 91% pass rate on business questions in a new Turing-built benchmark, outperforming frontier agents while finishing three to four times faster and using less than half the tokens. This performance is driven by Genie Ontology, which extracts critical business context directly from workspace notebooks, dashboards, and saved SQL queries.
Bring your business context to ChatGPT, Claude, Microsoft Copilot Cowork and Grok Bot
Databricks Genie now integrates natively with ChatGPT, Claude, Microsoft Copilot Cowork, and Grok Bot, alongside an MCP server connection for other AI agents. These external tools can draw on the Genie Ontology to run SQL over authorized data and return cited business answers directly into drafting, planning, or messaging workflows.