1 From automation to autonomy

If you are an enterprise architect, you are facing a challenge unlike any in the past two decades. By 2028, Gartner forecasts that 33% of enterprise software applications will incorporate agentic AI, up from less than 1% in 2024. These systems are not assistants waiting for your approval. They perceive context, reason through options, and act autonomously to achieve defined goals. They operate at machine speed, making decisions in milliseconds that would have taken your organization hours or days to approve through traditional governance processes.

1.1 Case study: when autonomy strikes at 3 a.m.

1.2 The quiet shift from automation to autonomy

1.2.1 The difference between automation and autonomy

1.2.2 Agentic AI: autonomy with perception and reasoning

1.2.3 Agentic AI and generative AI: Clarifying the distinction

1.2.4 Why copilots don’t solve the agentic architecture challenge

1.3 Why autonomy is no longer optional

1.4 What changes for enterprise architecture

1.4.1 A new mental model: blueprint plus control room

1.5 A simple example: manufacturing without and with a foundation for autonomy

1.5.1 How autonomy touches each architectural layer

1.6 Implications for enterprise architecture as a discipline

1.6.1 From a Foundation for Execution to a Foundation for Autonomy

1.7 How this book works

1.8 Summary