Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

Single Sign-On (SSO) means fewer password headaches, faster access, and better security for human users. But the same cannot be said for AI agents. SSO, a core part of Identity and Access Management (IAM), which was initially built for humans, can no longer be used for AI agents. For humans, it was quite simple - just log in once, and authenticate across connected apps. However, when an AI agent tries to authenticate the same way, the traditional access model breaks fast.

Zero-Shot Learning is a podcast about how AI gets built, secured, and deployed. Hosted by Nancy Wang, 1Password CTO, and Dev Tagare, Senior Director of Engineering at Google, it's a builder's view of the architecture and the complex choices it takes to ship with AI.

As AI becomes embedded in insider risk platforms, organizations are evaluating more than detection capabilities. Governance, data privacy, model control, and transparency are now central to the assessment of AI-enabled IRM solutions.

In my last post, I explained the math behind cosine similarity. Cosine similarity is a powerful search technique. When you are dealing with thousands or millions of chunks, it provides a fast, scalable way to find content conceptually similar to the user’s question. That is a major breakthrough. Without vector search, modern RAG would be much harder to build. But the mistake is pushing every retrieval problem into vector search. That is where practical retrieval starts breaking down.

Traditional API security protects deterministic systems with known endpoints and explicit actions, while MCP-powered AI agents operate through inferred intent, dynamic tool chaining, and natural language interactions. This requires MCP-specific security controls such as tool governance, behavioral monitoring, and semantic anomaly detection.

The first time a security team needs an AI agent audit trail is usually 72 hours after the agent has already done something it shouldn’t have. Detection fires. Someone pulls every relevant log from the SIEM (Kubernetes audit, container runtime, cloud audit) and three hours in realizes the events that actually matter were never written. Which prompt triggered the tool call. Which parameters the agent passed. Which output left the cluster.

Every AI-SPM tool runs posture and detection with a single arrow: runtime evidence flowing back to rank posture findings. The load-bearing direction runs the opposite way, and almost nothing runs it — posture flowing forward to tell the detection layer what an attack even looks like.