Secure your AI application transformation with Security
summary
Companies report 85% seeing AI benefits, 73% fearing new harms and 48% blocking AI, so this session shows concrete defenses: Microsoft Purview labels and prevents oversharing (live Copilot Studio demo returns labeled summaries and respects permissions), a preview Purview SDK for apps (summer target), model scans by HiddenLayer in the Azure AI catalog, Azure AI’s Prompt Shield blocking prompt‑injection in real time, and Defender for Cloud surfacing jailbreak/wallet alerts plus remediation scripts.
AI-written summary · about
chapters (5)
- 0:00Securing AI applications
- 7:21Microsoft Purview for data
- 24:22Azure AI Content Safety
- 29:26Defender for Cloud security
- 36:15Summary & Q&A
related talks
28mSecuring AI by design
01h 11mKeynote - Securing AI: Building Trust in the Era of AI
09mProtecting AI Agents with Microsoft Purview
42mSecuring the Invisible
37mReinforce AI security: Protecting AI applications, models, and data
15mAI Guardrails for Enterprise Deployments
39mThe Future of Cybersecurity in an AI-driven Era
18mCommit Secure AI: Embedding Security in Code, Pipelines, and Production