Mike O'Donnell Archives | STAGING

What is MCP in a Regulated Workflow? Four Governance Breakpoints Key Takeaways An MCP-driven workflow in a regulated environment can fail compliance in four distinct ways; identity attribution, audit completeness, data residency, and evidentiary chain. Identity attribution failure produces the other three as downstream symptoms. The Model Context Protocol specification, as of its 2026 roadmap, […]

MCP vs REST The Enterprise Decision Is More About Governance Than Integration Key Takeaways An MCP server is a software layer that exposes typed tools, resources, and prompts to AI clients through the Model Context Protocol, an open standard introduced by Anthropic in November 2024 to give LLM applications a uniform way to connect to […]

What is MCP? The Model Context Protocol Explained Key Takeaways The Model Context Protocol (MCP) is the broadly deployed open standard for connecting AI applications to external tools and data sources, released by Anthropic in November 2024. MCP standardizes one slice of the AI integration problem: tool discovery, invocation, and transport between an AI client […]

What is MCP for AI Agents? When the Protocol Is Enough, and When You Need the Layers Above It Key Takeaways Model Context Protocol (MCP) is an open standard that helps AI agents discover and call external tools. It solves one specific part of the agent stack well. MCP doesn’t handle memory between steps, long-running […]

What is an MCP server? Key Takeaways Model Context Protocol (MCP) server provides AI agents with access to your business’s tools, data, and systems. It does not automatically provide governance over that access. The conversation goes User → LLM → Agent → MCP Server → Enterprise Systems. Each layer has a different job. The MCP […]

Does MCP replace GraphRAG? Key takeaways MCP is a popular access layer to GraphRAG. It increases the accessibility of who can query the graph, while leaving the graph’s content and governance unchanged. Knowledge graphs encode meaning at ingest. MCP exposes that meaning at runtime to any compliant AI client. The bottleneck on most graph programs […]

What Is Contextual Intelligence? Key Takeaways Contextual intelligence is a data system capability that enables accurate reasoning by integrating five dimensions of context: semantic meaning, relational connections, temporal history, behavioral patterns, and operational constraints. Lack of contextual intelligence is one of the primary reasons AI projects fail. Gartner predicts that through 2026, organizations will abandon […]

What Is an Enterprise Knowledge Graph? The Governance Layer That Makes a Graph “Enterprise-Grade” Key Takeaways An enterprise knowledge graph is a knowledge graph built for enterprise deployment. It adds formal ontology governance, entity resolution across multiple source systems, role-based access controls, and provenance tracking to the core graph model. What makes the system ‘enterprise’ […]

What is an Ontology? A practical guide for enterprise data and AI teams. Key Takeaways An ontology is a formal, machine-readable model that defines what concepts mean, how they relate to each other, and what rules govern those relationships. Without a shared ontology, the same real-world entity appears under different names in every system: “customer” […]

What is a Context Layer, in simple terms? Key Takeaways Gartner’s Intelligence Capabilities Framework positions the context layer as a distinct architectural tier, sitting between the data layer and the intelligence layer. It acts as a translation layer that helps make enterprise data usable and governable for AI systems. A context layer maintains and exposes […]