- Models: The underlying AI systems that interpret prompts, generate responses, and make predictions
- Tools: The mixing layer that connects AI to enterprise systems, equivalent to APIs, protocols, and connectors
- Context: Before making decisions, information agents need to grasp the complete business picture, including customer histories, product catalogs, and provide chain networks
- Governance: The policies, controls, and processes that ensure data quality, security, and compliance
This framework helps diagnose where reliability gaps emerge. When an enterprise agent fails, which quadrant is the issue? Is the model misunderstanding intent? Are the tools unavailable or broken? Is the context incomplete or contradictory? Or is there no mechanism to confirm that the agent did what it was speculated to do?
Why that is an information problem, not a model problem
The temptation is to think that reliability will simply improve as models improve. Yet, model capability is advancing exponentially. The fee of inference has dropped nearly 900 times in three years, hallucination rates are on the decline, and AI’s capability to perform long tasks doubles every six months.
Tooling can also be accelerating. Integration frameworks just like the Model Context Protocol (MCP) make it dramatically easier to attach agents with enterprise systems and APIs.
If models are powerful and tools are maturing, then what’s holding back adoption?
To borrow from James Carville, “It’s the information, silly.” The basis explanation for most misbehaving agents is misaligned, inconsistent, or incomplete data.
Enterprises have amassed data debt over a long time. Acquisitions, custom systems, departmental tools, and shadow IT have left data scattered across silos that rarely agree. Support systems don’t match what’s in marketing systems. Supplier data is duplicated across finance, procurement, and logistics. Locations have multiple representations depending on the source.
Drop a couple of agents into this environment, and they’re going to perform splendidly at first, because each is given a curated set of systems to call. Add more agents and the cracks grow, as each builds its own fragment of truth.
This dynamic has played out before. When business intelligence became self-serve, everyone began creating dashboards. Productivity soared, reports didn’t match. Now imagine that phenomenon not in static dashboards, but in AI agents that may take motion. With agents, data inconsistency produces real business consequences, not only debates amongst departments.
