
4 minute read
Manufacturers don’t need another AI tool that reports what went wrong yesterday. They need one that understands what’s happening now, why it matters, and what to do next, without breaking the rules that keep the business running.
That’s the promise of Infor Industry AI for Claude, a new connector that links Claude with Infor’s manufacturing applications through Infor Velocity Suite. For organisations already using Claude, it means asking questions about operations in plain language, getting answers grounded in real manufacturing context, and taking approved action in Infor, all in the same conversation. The division of labour is simple: Claude brings the reasoning and the conversational experience; Infor brings the industry intelligence
Why context is everything
A general-purpose AI can summarise data you hand it. But ask whether a production order is at risk, whether it can still ship on time, and what you should do about it, and the question gets much harder.
A good answer depends on how everything connects: bills of materials, inventory and allocations, open purchase orders, machine and labour capacity, customer priorities, available-to-promise logic, site rules, and the permissions of the person asking. All of that already lives in Infor.
Rather than simply pointing an AI model at ERP data, the connector brings together Infor CloudSuite data, business objects, manufacturing logic, roles, entitlements and governance. Access to data gives you an answer. Understanding the business behind the data helps you choose the right action.
Recommendations that turn into governed action
The connector supports both reading from and writing to Infor. Depending on the use case and the user’s permissions, that can include creating customer orders, releasing purchase orders, issuing raw materials to production, shipping orders, and recording inventory or labour transactions.
Every action stays within the user’s existing roles, permissions and approval controls. When a decision could affect another order or customer, the agent comes back to the user for sign-off. And if there’s no viable way to meet a commitment, it says so and offers alternatives rather than forcing an unrealistic plan.
The guiding principle: agents do the legwork; people make the calls.
The bigger picture
This isn’t about putting another chatbot in front of ERP. It’s about closing the gap between understanding what’s happening, deciding what to do, and acting on it. That’s at the heart of Infor’s vision for the Agentic Enterprise: people and AI working together across connected processes, with intelligence grounded in industry context and actions kept within governed boundaries.
In the end, the value of AI in manufacturing won’t be measured by how well it talks about the business, but by how well it helps people run it.