Presentation
AI MCP Protocol Explained: The Right and Wrong Use Cases for MCP in z/OS
Contributor
Event Type
Technical Session
Enterprise Architecture
Machine Learning/AI
Network Security and Management
Soft Skills
All Audiences
Modernize Data Management and Analytics
TimeWednesday, August 199:15am - 10:15am EDT
LocationRoom 308
DescriptionMCP is generating a lot of buzz in the AI world, but what does it actually mean for mainframe professionals, and should you care?
The Model Context Protocol (MCP) is fast becoming a key building block in how AI systems connect to external tools, data sources, and services. But like any emerging standard, it comes with trade-offs — and on z/OS, the decisions you make about when and how to use it carry real consequences for performance, security, and maintainability.
This session cuts through the hype and gives mainframe professionals a clear, practical view of what MCP is, how it works, and most importantly where it fits and where it doesn't. We'll explore real-world scenarios where MCP is a genuine enabler for integrating AI capabilities with z/OS systems, and be equally honest about the situations where it's the wrong tool for the job.
No prior knowledge required. If you're evaluating AI integration options for your mainframe environment, or just trying to make sense of a term you keep hearing, this session will give you the grounding you need to make informed decisions.
What we'll cover:
- What MCP is and how it enables AI models to interact with external systems and data
- How MCP applies specifically to z/OS environments and mainframe data sources
- The right use cases — where MCP adds genuine value in a mainframe context
- The wrong use cases — where alternatives are safer, simpler, or more appropriate
The Model Context Protocol (MCP) is fast becoming a key building block in how AI systems connect to external tools, data sources, and services. But like any emerging standard, it comes with trade-offs — and on z/OS, the decisions you make about when and how to use it carry real consequences for performance, security, and maintainability.
This session cuts through the hype and gives mainframe professionals a clear, practical view of what MCP is, how it works, and most importantly where it fits and where it doesn't. We'll explore real-world scenarios where MCP is a genuine enabler for integrating AI capabilities with z/OS systems, and be equally honest about the situations where it's the wrong tool for the job.
No prior knowledge required. If you're evaluating AI integration options for your mainframe environment, or just trying to make sense of a term you keep hearing, this session will give you the grounding you need to make informed decisions.
What we'll cover:
- What MCP is and how it enables AI models to interact with external systems and data
- How MCP applies specifically to z/OS environments and mainframe data sources
- The right use cases — where MCP adds genuine value in a mainframe context
- The wrong use cases — where alternatives are safer, simpler, or more appropriate