From insight to action: Bringing Claude into the operations that run manufacturing

Manufacturers do not need more AI that simply tells them what happened. They need AI that understands what is happening across their operation, why it matters, what can be done about it, and how to act without stepping outside the rules that keep the business running.

That is the opportunity behind Infor Industry AI for Claude, connecting Claude with Infor's manufacturing applications through Infor Velocity Suite. For customers already using Claude, it provides a natural way to ask questions about their operations, receive answers grounded in manufacturing context, and take approved action in Infor from the same conversation.

The idea is simple: Claude provides the reasoning and conversational user experience. Infor provides the industry intelligence behind it.

AI is only as useful as the context behind it

Ask a general AI model whether an order is late, and it may be able to summarize the information you give it.

Ask whether a production order is at risk, whether it can still be delivered on time, and what action should be taken, and the problem becomes very different.

The answer depends on understanding the relationships across the operation: the bill of materials, inventory and allocations, open purchase orders, production capacity, labor, customer priorities, available-to-promise logic, site rules and the permissions of the person asking the question. That operational context already lives within Infor.

Infor Industry AI for Claude brings that context into the Claude experience. Instead of simply connecting an AI model to ERP data, the connector brings together Infor CloudSuite data, business objects, manufacturing logic, roles, entitlements and governance.

That distinction matters.

Access to data can give AI an answer. Understanding the business behind the data helps it recommend the right action.

The architecture behind the use cases connects Infor CloudSuite orders, inventory and allocations, BOMs and routings, machine and labor capacity through a shared knowledge layer, with specialist skills across production, inventory, buying and customer service.

One connection solving very real manufacturing problems.

The value becomes clearer when you start with the problems manufacturers deal with every day rather than with the technology.

Here’s what that looks like for three critical roles every manufacturer depends on: the production planner, customer service representative and buyer. Each starts with a different problem, but all three require AI to understand what is happening across multiple parts of the operation before recommending what to do next.

Help a production planner resolve an at-risk order before it becomes a late order

A planner asks a straightforward question:

Which production orders are at risk, and what can I do about them?

Answering it requires more than identifying a red flag.

Claude, using Infor's specialist skills, can examine open production orders, determine whether the risk comes from materials, machine capacity or labor, and work through possible responses. If material is short, it can check what has been issued, what is available, and whether incoming purchase orders will arrive in time. If capacity is the constraint, it can look for alternatives. If the problem cannot be resolved, it can recommend rescheduling and start the process of alerting customer service.

For the planner, the shift is from finding problems to resolving them.

Help customer service make a promise based on current Infor data

Now imagine a customer service representative receives a rush request while the customer is still on the phone.

Instead of calling planning, checking multiple warehouses and waiting for someone to confirm capacity, the rep can start with the customer's request.

Using data pulled directly from your Infor systems, Claude can check finished stock across locations, account for inventory already allocated to other customers, evaluate whether supply can be safely reallocated, and - when necessary and once the rep approves - create a production order and evaluate the materials and capacity required to fulfill it.

If fulfilling the rush request affects another customer order, the agent can surface that impact. The displaced order can be replanned and mitigation options presented for review. And if the requested date simply cannot be met, the agent can say so and recommend the best alternative.

The result is not simply a faster answer. It is a more confident customer commitment because the operational reality behind the promise has already been considered.

Help a buyer understand the true impact of a supplier delay

A supplier says a purchase order will arrive 20 days late.

The obvious question is not just where is my PO?

It is:
What does this delay mean for production and for my customers?

Working that out manually can mean tracing the material through production orders, subassemblies, finished goods and customer deliveries.

Again, using Infor data, Claude can analyze those dependencies, identify affected production and customer orders, evaluate alternative materials and suppliers, and propose actions such as creating an expedited purchase order for the immediate requirement while keeping the original order open for remaining demand.

That gives the buyer something more valuable than an alert: an understanding of the business impact and a path to recovery.

From recommendation to governed action

This is where the story moves beyond conversational AI.

Infor Industry AI for Claude supports read and write interactions with Infor. Depending on the use case and the user's permissions, that can include actions such as creating customer orders, releasing purchase orders, issuing raw materials to production, shipping orders and performing inventory or labor transactions. Actions remain subject to the user's existing roles, permissions and approval controls.

The principle is straightforward:

Agents do the legwork. People make the calls.

When an action could affect another order or customer, the agent can return to the user for approval. Findings are grounded in live Infor CloudSuite operational data, and if the system cannot find a viable way to meet a commitment, it can surface the limitation and present alternatives rather than force an unrealistic plan.

Open by design, governed by Infor

Manufacturers rarely operate in a single technology environment. Their people move between ERP, supply chain, productivity and collaboration tools throughout the day.
Infor Industry AI for Claude reflects that reality.

Built using the open Model Context Protocol, or MCP, the connector gives customers who already use Claude a supported way to reach Infor from the experience they have chosen. The connector is built by Infor, while customers maintain Infor's existing roles, permissions and approval controls.

And importantly, Claude does not replace Infor Velocity Suite.

Infor Velocity Suite remains the foundation for Infor's industry intelligence, governance, ontology, precision and industry use cases. Customers can continue to use Infor's GenAI Assistant, role-based workspaces and guided experiences. The Claude connector provides another way for Infor customers who already use Claude to access that depth. Infor Industry AI for Claude is available to select Infor Velocity Suite customers. It requires an eligible Claude plan. On Claude Team and Enterprise plans, an Owner must enable the connector first.

Another step toward the Agentic Enterprise

The bigger opportunity is not about putting another chatbot in front of ERP.

It is about closing the distance between understanding what is happening, deciding what to do, and taking action.

That is central to Infor's vision for the Agentic Enterprise: people and AI working together across connected business processes, with intelligence grounded in industry context and actions operating within governed boundaries.

Getting there means meeting people where they already work. Manufacturers run many systems, and their teams have their own favorite tools. Our commitment is to stay open and connected to that reality, and to make Infor easy to use from wherever work happens. That could be Claude, an Infor workspace, or both in the same day.

Infor Industry AI for Claude brings that idea into tangible manufacturing moments: a planner trying to keep production on schedule, a service representative trying to make a promise with confidence, or a buyer trying to protect customer commitments when supply changes unexpectedly.

Because ultimately, the value of AI in manufacturing will not be measured by how well it can talk about the business.

It will be measured by how well it helps people run it.