What Are MCP Connectors and
Why Do They Matter for Manufacturers?

Artificial intelligence is rapidly changing the way businesses interact with information. Past sci-fi films alluded to ‘intelligence’ being asked about a subject and then providing the information – an example is Star Trek (yes really and yes I am a fan) where the likes of Captain Kirk would ask the Starship Enterprise “computer” to find out the status of the “away team” or what was “happening in engineering”, well this is now possible with your own earth based system and company! Read on to find out more……

Tools such as ChatGPT and Claude can already write documents, analyse spreadsheets and help people make sense of complex information.

The next stage is potentially much more significant: allowing AI assistants to securely interact with the business systems companies use every day.

One of the technologies helping to make this possible is MCP Connectors, based on the Model Context Protocol (MCP). These connectors provide a standardised way for AI applications to interact with external systems, tools and sources of information.

What is MCP?

MCP is an open standard that allows AI applications to connect with external systems, tools and sources of information.  A simple way to think about it is as a standardised bridge between an AI assistant and business software.

Without a connection, ChatGPT or Claude does not automatically know what is happening inside your ERP or manufacturing system.  With an appropriately configured MCP connection, an AI assistant can potentially access permitted information from that system and use it to answer questions.  Instead of logging into an ERP system, navigating through menus, finding a report, applying filters and interpreting the results, a user could simply ask:

“Which manufacturing orders are overdue?”  or: “Which materials are currently below safety stock?”

The AI assistant can then use the connected system as a source of information when answering the question.

What could this mean in practice?

Consider a manufacturing business using an MRP or ERP system to manage inventory, sales and production.  A production manager might normally need to access several screens to understand what’s happening.  An AI assistant connected to the system could potentially allow them to ask questions such as:

✅ Which customer orders are due this week?
✅ Which manufacturing orders are running late?
✅ Which materials could delay production?
✅ What products are below safety stock?
✅ Which customers have the highest value of open orders?

This creates a very different way of interacting with business software.  From software users to business conversations. For decades, business software has largely required people to learn how the software works.  Users learn menus, dashboards, reports, filters and terminology.

AI could begin reversing that relationship.

Instead of teaching employees how to find information inside a system, the system increasingly understands the question being asked in everyday language.  That could be particularly valuable for smaller manufacturers where managers often perform several roles and may not spend enough time in the ERP system to become expert users.

MCP doesn’t mean unrestricted AI access

Connecting an AI assistant to business software should not mean giving AI unrestricted access to everything.  Security, authentication, user permissions and appropriate controls remain extremely important.

Different connections can also provide different capabilities. Some may allow information to be queried, while others may allow authorised actions to be performed.  Businesses therefore need to consider not only what AI can do, but what AI should be allowed to do.

Why this matters for manufacturing

Manufacturers generate enormous amounts of operational information:

✅ Inventory
✅ Sales orders
✅ Purchase orders,
✅ Manufacturing orders,
✅ BOMs
✅ Customers
✅ Suppliers
✅ Lead times
✅ Production data.

The challenge has traditionally been turning all of that data into useful information quickly enough to make better decisions.

MCP and AI assistants could help close that gap.

The future of manufacturing software may therefore be less about creating another dashboard or report and more about simply being able to ask your business a question, bringing Captain Kirk and sci-fi possibilities closer to current reality!

How Langcroft Consultancy can help

As AI connectivity develops, we believe technologies such as MCP will become an increasingly
important part of the connected manufacturing technology landscape.