Part 1 — Foundations of Generative and Agentic AI

Overview

Before an organisation can deploy Agentic AI responsibly, it needs to understand what it is building on. Part 1 establishes the technical, economic, and organisational foundations that underpin everything that follows in this book.

We begin with history — tracing the arc from the earliest statistical language models through the rule-based chatbots of the 1960s and 1990s, to the transformer-powered systems that now handle billions of conversations daily. This history is not merely academic. It explains why today’s systems behave the way they do, what limitations are inherited from earlier paradigms, and what genuinely changed with the arrival of large language models.

From there, Part 1 examines three defining practical realities of the current moment: the economics of running AI at scale, the emergence of multimodal systems that can interpret text, images, audio, documents, and code, and the ways generative AI is already creating value inside organisations.

Together, these chapters explain the transition from generative AI as a responsive tool to Agentic AI as an operating capability — systems that can reason across context, use tools, participate in workflows, and act under human-defined constraints.

---

Chapters in This Part

Chapter	Title	Theme
1	From Statistical Models to Intelligent Systems	History and foundations
2	The Economics of AI: Capability, Speed, and Cost	Economics of deployment
3	How AI Learned to See, Hear, and Read	Multimodal AI
4	Generative AI in Practice: Where Enterprise Value Is Created	Enterprise adoption and value creation

---

Part 1 is the recommended starting point for readers who want the full conceptual foundation before moving into architecture, deployment, risk, and governance.