0:00
/
0:00
Transcript

The Next Big Thing in AI: Model Context Protocol (MCP) and Agentic AI

A deep dive into the cutting-edge technology that's unlocking AI agents

Hi everyone! Kelly and I are taking some time this summer to relax and reflect on the rapidly evolving world of AI and education. Today, I want to introduce you to a concept that's brand new and incredibly exciting - something we're all still learning about together.

It's called Model Context Protocol, or MCP for short.

What is Model Context Protocol (MCP)?

An MCP is a special kind of API that conforms to a specific format, allowing large language models to use that API in really neat and interesting ways.

Think about the last time you used an AI system and asked it to look up information about the latest sports scores. Maybe it was able to figure that out by doing a search on the internet and then using that search as a prompt to feed back into the model.

But MCPs are even cooler than that. They provide a structured way for models to go get information from various sources on the internet and have that information be accurate and up-to-date. So instead of the AI guessing at what the sports score is, it's getting the real information from that endpoint in a way that the model can understand it, process it, and use it in a much richer way.

Learn more about MCP:

The Power of Agentic AI

What's really interesting about this is that MCP also unlocks something called agentic AI - where the AI can act as an agent on your behalf. This is possible because MCPs also define actions that the model can perform on your behalf in a structured way.

Since I'm traveling and I've flown this week, let's talk about another kind of agent: a travel agent.

The Travel Agent Example

Think about the interaction you have with a travel agent when booking a vacation. You might call the agent and ask them:

  • Information about what trips are available

  • What the airfare to your favorite destination looks like

  • What are the specials this week

That's an example of an MCP interaction where the model might be trying to retrieve information from that endpoint and use it to give you a better response.

But when you go to the next level with an MCP, you can also define actions that it can take. So not only can the MCP provide information about those airfares and flights, it also gives the model the opportunity to book those flights on your behalf and give you a confirmation number in return.

Explore Agentic AI further:

Why This Matters Right Now

These MCPs are really powerful, and they're brand new. We're seeing a lot of new places that are bringing them out. This is the latest trend, unlocking this whole power of agentic AI.

Being able to do these things represents some of the latest cutting-edge stuff in artificial intelligence. Gartner has identified agentic AI as the top strategic technology trend for 2025, and we're seeing rapid adoption across industries.

Hands-on Learning Resources:

The Big Picture

Keep an eye on this space. Look for the MCP term. Look for agentic AI. Recognize that it's really just a special kind of API that you can even write yourself.

The exciting part is that we're at the very beginning of this technology. It's so new that we're all learning how it works together. But the potential is enormous - we're talking about AI systems that don't just answer questions, but can actually take meaningful actions in the real world on your behalf.

Whether you're an educator, developer, or just someone curious about where AI is heading, now is the perfect time to start exploring these concepts. The foundations are being laid right now for what could be a fundamental shift in how we interact with artificial intelligence.

More on this to come as we continue exploring the intersection of AI and education!


Want to stay updated on the latest developments in AI and educational technology? Subscribe for more insights as we navigate this exciting frontier together.

Discussion about this video

User's avatar