BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//hacksw/handcal//NONSGML v1.0//EN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTAMP:20260722T235918Z
DESCRIPTION:Click for Latest Location Information: http://dgiq-aigov2026.da
 taversity.net/sessionPop.cfm?confid=166&proposalid=16771\n\nAs organization
 s scale generative AI and retrieval-augmented generation (RAG), they encoun
 ter familiar challenges: inconsistent outputs, uncontrolled document sprawl
 , and difficulty sustaining trusted AI beyond initial pilots. This session 
 introduces a practical approach to accelerating and maintaining enterprise 
 AI by leveraging an existing but underutilized capability—data govern
 ance.\n\nI'll show how governed assets—cataloged data, linea
 ge, data quality controls, and curated documentation—can be transform
 ed into a scalable foundation for domain-specific RAG agents. Central to th
 is approach is a governance-driven operations pipeline, where data stewards
  and subject matter experts curate, validate, and continuously improve AI-r
 eady knowledge stores.\n\nRather than relying on a single monolithic model,
  this framework enables a multi-agent architecture grounded in a trusted, d
 omain-specific context. The result is more precise, explainable, and sustai
 nable AI that aligns with enterprise standards.\n\n\n
 How to accelerate RAG adoption using existing governance artifacts and proc
 esses\n
 Designing a governance-driven pipeline for curating and maintaining AI-read
 y knowledge\n
 Leveraging data stewards and SMEs to ensure accuracy, relevance, and sustai
 nability\n
 Why domain-specific agents outperform monolithic RAG approaches at scale\n
 Embedding lineage, data quality, and policy controls to improve trust and e
 xplainability\n
 Establishing continuous improvement loops for long-term AI performance and 
 governance\n
 Creating an environment ready for higher-level AI use cases using orchestra
 tion\n\n\n
DTSTART:20261117T164500
SUMMARY:How Existing Governed Assets Can Be Leveraged for Domain-Specific R
 AG Agents
DTEND:20261117T172959
LOCATION: See Description
END:VEVENT
END:VCALENDAR