Presented by SAP
Enterprise AI agents that share the same data can still make conflicting decisions because raw data rarely captures the business context — the processes, policies and priorities that determine what that data means to the company. As agents take on more decision-making, those gaps in context become harder to contain.
Three problems emerge when multiple agents act without shared business semantics: identical figures lead different functions to different conclusions, agents that work correctly on their own combine into the wrong outcome, and accuracy degrades as data accumulates.
Enterprises have spent years loading ever more records into central data platforms, separating them in the process from the policies, processes and business context where they originated. That separation is at the root of the problem, says Irfan Khan, president and chief product officer for SAP Data & Analytics.
“If you access data from a source environment and it doesn’t have the context, the likelihood is you’re going to make an ill-informed decision,” he explains. "Users learn from history only when institutional knowledge helps them separate signal from noise. An agent starts with no such knowledge unless the company supplies it."
The agent coordination gap in multi-agent AI systems
Traditional data strategies extract records from source systems and centralize them, which separates the data from the semantics and process context it carried at the source. The shift toward separating storage and compute let companies scale each independently and encouraged many to build multiple data lakes alongside on-premises estates, flat files and object stores, leading to fragmented data increasingly separated from its original business context.
In multi-agent systems, missing context doesn’t stay a data problem. Once agents divide a business process among them, it becomes a coordination failure. For instance, a procurement agent negotiates longer payment terms to improve margins, a treasury agent delays payments to preserve liquidity, and a compliance agent holds payments until every validation clears. Each one is performing its assigned job correctly, but together they're stalling supplier payments long enough that critical vendors start deprioritizing the company, and no internal system flags anything as broken.
Plus, few business processes run on a single agent from end to end. An order entry agent might hand off to one that checks inventory, which hands off to another that files a procurement request or decrements stock. If those steps run out of sequence, one agent can decrement an account before another has confirmed the goods exist, and the task can be completed, but the outcome can still be wrong. And there's no reason agents would coordinate any better than people do when nobody manages the handoffs, Khan says.
“It’s almost like pushing out a task to 20 individual people to go and get it done for a client, and then you realize that nobody held a team meeting,” he says. “Nobody kept an inventory of where you were, or where you are within the process.”
A business data fabric gives agents shared business context
Giving agents access to distributed enterprise data solves only part of the problem. They also need a common understanding of how the business expects that data to be interpreted and acted upon.
SAP’s business data fabric architecture is designed to preserve that meaning as data moves across systems rather than separating it from its original business context. It connects data and its semantics with the processes, policies, and other business knowledge that govern how the data is used, while allowing agents to reach information across distributed environments.
That shared context includes what Khan describes as “company memory”: the KPIs, operating practices and accumulated business knowledge that a company wants its agents to apply consistently.
“It’s almost like having a village elder who has all the historical understanding, and you would consult that village elder to make an informed decision,” Khan said. “There isn’t such a thing in the agentic world, so let’s consider that to be company memory.”
The point is to give agents a common frame of reference rather than allowing each to optimize its own piece of a workflow in isolation. An inventory agent, for example, shouldn’t make the locally optimal decision if that decision conflicts with a broader business objective or policy. The shared semantic and policy layer gives each agent a way to determine whether it is still on the right execution path — and when it should fall back rather than act.
That requires encoding constraints as well as objectives. Agents need to understand not only what they are trying to accomplish, but what they must never do and which conditions must be satisfied first, such as confirming goods are available before authorizing a shipment. Within SAP’s business AI platform, the agent harness and agent hub provide the control layer for defining those permitted actions and constraints. The goal is to prevent an individually rational agent decision from producing an outcome that violates policy, governance requirements or the objectives of the larger process.
The business context agents need already exists
The good news is that enterprises don’t have to create all of this context from scratch. Much of it already exists in the systems they use to run the business. Business processes, master data, operational rules and governance policies represent years of accumulated knowledge about how a company actually operates.
The challenge is making that knowledge available to agents alongside the underlying data. For SAP customers, Khan argues, decades of business processes, master data and operational rules already reside in their enterprise systems. Business Data Cloud is designed to connect that existing context across systems and applications so it can travel with the data agents use.
“It’s all well and good now being able to access data in a very free-flow, self-service fashion, and a prompt can access pretty much any of your data,” he says. “But what travels with the context is the security, the governance and the master data.”
And that context, Khan argues, will increasingly separate AI experimentation from business value.
“Nobody has a shortage of data, but the contextual data is what really is the prize,” Khan says. “Making sure that you’re grounded in that context is going to be the difference between being successful in enterprise AI and just being part of what takes place today, where everyone is experimenting but finds no real appreciable value.”
Sponsored articles are content produced by a company that is either paying for the post or has a business relationship with VentureBeat, and they’re always clearly marked. For more information, contact sales@venturebeat.com.