Bruno Kurtic, Bedrock Data | RSAC 2026
summary
Claims that AI security is primarily a data problem are justified with concrete artifacts: a crawling metadata lake that inventories petabytes across on‑prem, cloud and SaaS; an ARGUS AI risk model defining risk as a triad of data sensitivity, entitlements and exposure, and agent capabilities; a released MCP server to enforce agent access; and an AI DBOM that maps PII/PHI, business taxonomies and bias signals. Cites a Fortune 150 financial win and the 15% SOC‑alert triage stat to quantify impact.
AI-written summary · about
chapters (7)
- 0:00Intro
- 2:42Enhancing Data Management: AI Solutions and the Power of Metadata
- 5:20Building Seamless Data Solutions: From Metadata Lakes to SaaS Integration
- 8:19Collaboration with Snowflake
- 10:47Collaborative Innovations in Open Source and SIM Technologies
- 16:03AI Governance and Creating a Data BOM
- 18:53Strategic Expansion: A Comprehensive Outlook
