Business Central AI Agents: Your New Digital Coworkers

Every so often, a new coworker shows up and changes how work gets done. In the case of Business Central AI agents, the new coworker doesn't need a desk, won't touch the breakroom coffee, and is perfectly happy taking on some of the repetitive work nobody was exactly fighting over.

That sounds promising. It can also sound a little unsettling.

I've spent more than 20 years working with Dynamics, both as a consultant and on the client side leading IT and Field Service operations.

One thing I've learned is that when technology starts taking on work people used to do themselves, the first question usually isn't, "How sophisticated is the AI?"

It's more like, "Okay...what exactly is this thing going to do?"

That's the question I want to answer in this first article in a new series about AI agents in Dynamics 365 Business Central.

We'll look at what Business Central AI agents are, how they differ from Copilot and traditional automation, which agents are available in Business Central v28, where they could help with everyday work, and why people still remain firmly in control.

What are AI agents in Business Central?

Business Central AI agents are AI-powered capabilities that can work through multiple steps toward a defined business goal, interact with information in Business Central, and prepare work for employees to review.

Think of an agent less like a tool waiting for you to tell it what to do next and more like a coworker you've given a specific assignment.

You define the job and the boundaries, and the agent can work through several of the routine steps involved in getting that job done.

For example, an agent might monitor incoming information, identify the relevant records, extract or validate data, and prepare a transaction for review.

Instead of an employee handling every administrative step individually, the agent can move the work forward until it reaches a point where human attention is needed.

That's what makes Dynamics 365 Business Central agents interesting to me.

The goal isn't to remove people from the process. It's to free them from some of the repetitive work so they can focus on the parts that actually need their knowledge and judgment.

AI agents vs. Copilot vs. traditional automation: What’s different?

Traditional automation follows predefined rules, Copilot generally helps a person complete a specific task, and an AI agent can work through multiple steps toward an assigned outcome, bringing a person in when review or judgment is needed.

The easiest way I think about the difference is how much direction each one needs along the way.

Traditional automation is very good at “when this happens, do that.”

Copilot is more collaborative. You ask for help, and it assists with the task in front of you.

An agent can be given a goal and then work through a series of steps toward that goal within the boundaries you've established.

That doesn't make one approach better than the others. They solve different problems, and businesses may have good reasons to use all three.

That's also why understanding Business Central Copilot and agents matters.

The interesting part of agentic AI in ERP isn't simply that AI can help an employee complete a task.

It's that an agent can keep the work moving through several routine steps before asking that employee to step back in.

Meet the Business Central v28 agents

So what does all of this look like inside Business Central?

With v28, Microsoft is expanding the agent lineup into specific jobs businesses can recognize.

Sales Order Agent

The Business Central Sales Order Agent can monitor incoming customer requests, identify customers and products, prepare quotes and orders, and communicate with customers along the way.

Payables Agent

The Business Central Payables Agent can monitor incoming invoices, extract information, identify vendors, suggest classifications, and prepare draft purchase documents for review.

Expense Agent

The Expense Agent can turn receipts and other expense information into draft expense reports while keeping review and approval with the employee.

Agent Designer

Agent Designer takes a different approach. Instead of addressing one predefined process, it opens the door to custom agents built around specific business processes.

We'll come back to that idea later in this series, when we look more closely at how agents could support the way your own business works.

For now, the important point is that Business Central AI agents aren't one all-purpose AI assistant. They're designed around specific jobs and outcomes.

Do AI agents replace employees? Why humans still stay in control

No. Business Central AI agents are designed to handle repetitive administrative work while employees remain responsible for approvals, exceptions, controls, and the business decisions that require judgment.

That's an important distinction. An agent may be able to prepare the work, but that doesn't mean we're handing over accountability along with the data entry.

Microsoft makes that boundary particularly clear with its Expense Agent. The agent can extract information, categorize expenses, and prepare draft expense reports, but it doesn't post transactions or make permanent financial changes without human review and approval.

The same principle matters whether we're talking about accounting, sales, or Field Service. Technology can surface information and move routine work forward. But when something falls outside the norm, people still bring the context and judgment to decide what happens next.

After all, automating a bad decision only gets you to the wrong answer faster.

Where can Business Central AI agents help most?

Business Central AI agents are best suited to repetitive, process-driven work: data entry, document processing, matching and validation, routine customer or vendor follow-up, and preparing transactions for review.

The common thread is work that needs to get done but doesn't always need an employee's full attention at every step.

Gartner describes a similar shift in its discussion of the human-AI workforce, encouraging businesses to think less about what work AI can take away and more about the capacity it can create.

That's the opportunity I find more interesting. If an agent handles some of the routine steps, employees can spend more of their time solving problems, helping customers, and improving the business.

How should businesses begin?

Start with one clear, repetitive process where Business Central AI agents could remove administrative steps without removing the human judgment the process needs.

Before choosing an agent, make sure the underlying process and data are ready. Decide where employees need to review the work, where approvals belong, and what should happen when something falls outside the normal path.

AI works much better when we don't ask it to untangle a process we haven't untangled ourselves.

McKinsey makes a similar point in its guidance on change management in the age of AI, recommending that businesses choose the right processes to automate and involve employees in redesigning how the work gets done.

Start small. Solve a real problem. Then learn from it before giving your new digital coworkers a longer to-do list.

The goal isn't to automate every role or hand every decision to AI. It's to let technology take on more of the repetitive work so people have more time for the decisions, customers, and improvements that need their attention.

Understanding what agents can do is the first step. Next, we'll look at where that help can translate into measurable time savings and financial value. Then we'll explore how standard and custom agents could support the processes that make your business unique.

If you're wondering where an agent might make sense in your own Business Central environment, let's talk.

Sandlapper can help you identify a practical first use case and determine whether your processes and data are ready for it.

And whether you're exploring Business Central AI agents, evaluating a move from Dynamics GP, or simply trying to get more from Business Central, I host open Office Hours every Tuesday at 11:30 ET.

Bring your questions and let's talk through what makes sense for your business.

About the Author

Chris Boling is a founding partner of Sandlapper Dynamics, where he helps businesses streamline operations, enhance productivity, and achieve strategic growth through Microsoft Dynamics 365. With over two decades of experience in the Dynamics community, Chris combines deep technical expertise with a customer-first approach to guide organizations through digital transformation.

His unique perspective, shaped by years as both a consultant and an end-user, enables Chris to deliver practical insights that bridge the gap between technology and business outcomes.

Chris brings authenticity, empathy, and a commitment to sustainable growth to every engagement.

You can reach Chris on LinkedIn.