Episode 111 - April 17th, 2025 - [Securing Agentic AI]
In this episode Michael and Sarah talk with guest Amanda Minnich
(
)
about securing agentic AI systems, and the new security challenges they face.
We also discuss Azure Security news about Azure File Sync, Docker support in Azure and a new series of Secure Future Initiative videos with appearances from Michael, Sarah, and various guests.
We also discuss Azure Security news about Azure File Sync, Docker support in Azure and a new series of Secure Future Initiative videos with appearances from Michael, Sarah, and various guests.
Links
- Transcript
- Secure Future Initiative: Tech Tips
- Azure File Sync support for managed identities
- Retirement: Docker Content Trust on March 31, 2028
- Michael joking about Bogosort
- Zero Trust Agent (ZTA) Framework for Multi-Agent Collaboration
- AI Agents Under Threat: A Survey of Key Security Challenges and Future Pathways
- Security of AI Agents
- Context is Key for Agent Security
- AGENTPOISON: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases
- Technical Blog: Strengthening AI Agent Hijacking Evaluations
- Agentic AI - Threats and Mitigations
- Microsoft Responsible AI Standard, v2
- Agentic Autonomy Levels and Security
- MemGPT: Towards LLMs as Operating Systems
- Magma: A Foundation Model for Multimodal AI Agents
- Magentic-One: A Generalist Multi-Agent System for Solving Complex Tasks
- AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases
- Agentic Systems: A Guide to Transforming Industries with Vertical AI Agents