NVIDIA AI Podcast How AI Data Platforms Are Shaping the Future of Enterprise Storage - Ep. 281
110 snips
Nov 18, 2025 Jacob Liberman, Director of Enterprise Product Management at NVIDIA, dives into how GPU-accelerated storage transforms enterprise data management. He explains the challenges of integrating AI agents in production and the critical role of secure data access. Liberman highlights the importance of making unstructured data AI-ready, discusses the security risks of data copies, and illustrates how NVIDIA's AI Data Platform enables continuous processing and efficient workflows. Anticipate the evolution of storage solutions through innovative partnerships and real-world implementations.
AI Snips
Chapters
Transcript
Episode notes
Data Velocity Forces Continuous Processing
- Data velocity (growth plus change) forces continuous reprocessing to keep AI representations current.
- If you can't detect changed items, organizations end up re-indexing everything, wasting resources.
Dishwasher Analogy For Reindexing Pain
- Jacob Liberman uses a dishwasher analogy to explain unnecessary rework when you can't track which data changed.
- He compares re-indexing all data to rewashing every dish when you forget which one was dirty.
Copies Increase Security And Governance Risk
- Copying data for AI pipelines increases attack surface and breaks linkage to the source of truth.
- Liberman warns that disconnected copies can retain access even after permissions change, creating security risk.