GenAI governance in practice
summary
Claims that only 5% ROI of GenAI pilots succeed, shows up in a small data quality experiment that compares 10–50 documents using similarity searches and contradiction checks. Implements an internal stack: Open Web UI for local prototyping, an API gateway (Azure/AWS/Vertex AI) for policy-as-code, a Chrome-extension UI and React Genie knowledge agent that returns resumes and availability. Prototype outperforms a SaaS rival functionally, with scalability still unresolved.
AI-written summary · about
chapters (12)
- 0:00GenAI governance intro
- 1:10GenAI project success rates
- 2:37Three pillars of governance
- 4:21Self-service prototyping
- 6:53Integrating GenAI workflows
- 9:05Centralizing model access
- 10:58Curated knowledge bases
- 13:08Detecting data quality issues
- 17:28LLM ops and monitoring
- 20:21Golden data sets
- 23:10Governance maturity level
- 25:38Q&A
related talks
31mGenAI after the Hype: Transforming Organizations with GenAI-based Agents
20mGenAI agents - Evolution of automation
48mOperationalizing AI Across the Enterprise
49mWhere GenAI projects fail & how can leaders prevent it
20mBeyond the Buzz: Discovering High-Impact Use Cases in the Generative AI Era
28mIs GenAI the technology revolution we need? How can we harness it at work?
23mScaling GenAI Applications and Agentic AI for Business Competitive Differentiation
27mSoftware engineering with GenAI