For decades, enterprise databases have evolved in response to infrastructure shifts: from spinning disk to flash, from on-premises to public cloud. But the rise of AI introduces something fundamentally different. This isn’t a change in storage medium or deployment model. It’s a change in who … or what … is consuming the data.
AI systems don’t want reports. They don’t tolerate latency. They don’t operate in batches. They expect immediate access to operational truth. As inference engines and AI agents move closer to systems of record, the database is no longer just storing business state — it is becoming part of the reasoning path itself.
Inference is now a first-class database workload.
But the implications run deeper than access patterns and query load. Enterprise architecture was built around a human actor whose presence provided properties that nobody designed, nobody specified and nobody priced – natural rate limiting, implicit intent behind every transaction, accountability anchors, governance assumptions, capacity models, cost ceilings. Agentic AI is the first technology capable of removing those properties simultaneously, across every architectural domain at once.
This series examines what that means for the systems of record that enterprise organisations depend on – and for the people responsible for them.
Index
Part 1 – The Shift: How AI agents change the way systems of record are accessed, queried, and used.
Inferencing Is a Database Problem Disguised as an AI Problem
Databases Were Built for Humans – AI Agents Change the Equation
AI Doesn’t Read Dashboards… and That Changes Everything for Databases
Why We Spent 20 Years Protecting Databases from Analytics (and Why AI Just Broke That Truce)
The Cloud Is Built on Uniformity. Databases Are Not
The Hidden Cost of Letting AI Agents Query Live Systems
AI Can Replace Interfaces, Not Systems of Record
Part 2 – Under Pressure: How AI agents are dismantling the implicit protections that enterprise data architecture was never designed to make explicit.
AI Agents Don’t Just Add Load. They Change Its Shape
Transactions Assume Intent. Agents Don’t Guarantee It
Your Database Doesn’t Know What an Agent Is
The Brake Was Human. Now It’s Gone
The Audit Trail Was Your Ground Truth. It Isn’t Anymore
Your Database Was Sized for Humans. The Bill Arrives When Agents Connect
The Application Layer Used to Protect You. Now It Can’t
You Won’t See Failure First. You’ll See Cost
The People Driving AI Don’t Own the Systems It Overwhelms
Decades of Change Management. None of It Was Built for Agentic AI
Conclusion: What agentic AI revealed and what must now be built deliberately.
Enterprise Architecture Quietly Depended on Humans in Far More Places Than We Realised
Something New Has to Sit Between Agents and Databases
flashdba 