Trusted AI at Scale: Secure Governance and Scalable Management for Your AI Models
summary
Claims a single platform that treats every model as a first-class artifact and shows a working product: JFrog AI Catalog that unifies internal packages, Hugging Face models and external APIs, offers discovery, evidence-based governance, one-click serving with a JFrog token gateway and runtime monitoring, and a scanner that flags malicious models. Upcoming shadow AI finds models and API calls inside artifacts and Docker images. It also surfaces NIMs — containerized model+runtime bundles with ~400 tuned packages, plus Neotron claims of ~6x throughput and open 9 trillion-token data.
AI-written summary · about
chapters (6)
- 0:00Introduction and challenges
- 3:11Introducing JFrog AI Catalog
- 10:21Model discovery and governance
- 17:26Model serving and runtime
- 22:43NVIDIA NIMs partnership
- 31:16Summary and next steps
related talks
33mSecuring the AI Surge
28mSecuring AI by design
35mBuilding Enterprise Solutions in the AI-Native Era
15mEnterprise AI Runs on Trusted Data
37mReinforce AI security: Protecting AI applications, models, and data
47mEmbedding AI: Don’t Burn Your Fingers in the Heat of the AI revolution
43mMaking AI Resources Easy to Find, Use, and Connect
30mSecureAI: A Scalable, Secure, and Compliant AI Solution