Agentic Access Management: How to Secure AI Agents & Non-Human Identities скачать в хорошем качестве

Agentic Access Management: How to Secure AI Agents & Non-Human Identities

AI agents are logging into SaaS apps and modifying production environments faster than security teams can track. Traditional identity tools cannot govern this new wave of autonomous non-human access. In this video, we introduce Oasis Agentic Access Management (AAM™), the industry's first intent-aware infrastructure for securing autonomous AI agents. As organizations deploy agents from providers like OpenAI and Anthropic, they create a chaotic web of over-privileged identities, shared tokens, and hardcoded secrets. You will learn: 🔸The Risk: How shared tokens and hardcoded secrets in AI agents put production environments at risk. 🔸The Solution: How Oasis AAM™ infers agent intent, enforces deterministic policies, and issues ephemeral Just-In-Time (JIT) identities. 🔸The Demos: Watch live examples of securing a Sales Agent (Salesforce/SendGrid) and a Research Agent (Gemini/HubSpot) to prevent unauthorized bulk actions. Timestamps 0:00 - The New Attack Surface: AI Agents 0:27 - Introducing Oasis Agentic Access Management (AAM™) 0:40 - The 3 Pillars: Intent, Policy, and JIT Identity 0:58 - Demo 1: Governing a Sales Agent (Blocking Bulk Emails) 1:36 - Demo 2: Securing a Research Agent (Ephemeral OAuth Apps) 1:57 - Recap: Unified Visibility & Audit Proof Resources Explore Agentic Access:https://www.oasis.security/agentic-ac... Book a Demo: https://www.oasis.security/demo Follow Oasis Security on LinkedIn:   / oasissecurity   #AISecurity #AgenticAI #NonHumanIdentity #AgenticAccessManagement #OasisSecurity