AI in plain language. Six short reads.

Three types of AI. Two structural reads. One framework that holds them together. Each ends with a question to take into your next IT review. Or your next board meeting. Same idea, different chairs.

01 · Traditional AI

"The Analyst." The oldest, most mature form.

Learns patterns from history. Predicts what comes next, or classifies what just happened. Deterministic, auditable, narrow. Powers most of your existing BI today.

Read the insight
02 · Generative AI

"The Creator." Current centre of gravity.

Trained on text, image, code. Produces fluent new content from a prompt. Where every brand-name AI tool sits today. Strong, fluent, and prone to confident nonsense.

Read the insight
03 · Agentic AI

"The Worker." LLM + loop + tools.

Observes, decides, calls a tool, repeats until done. Powerful and early-stage. Governance-critical. Never give it autonomous reach into HR, finance, or customers.

Read the insight
04 · Before the four properties

How generative AI gets its character.

Built in two stages. Pretraining makes a document completer that has never heard of you. Fine-tuning overlays the helpful assistant. Fluent prose and confident nonsense come from the same machine.

Read the insight
05 · The 6-Layer Stack

From agent to silicon.

Agent, Orchestration, Inference, Transformer, Training, Infrastructure. Six layers, each with its own cost profile and its own strategic decision. Top-down: where the user sits. Bottom-up: where the spend lives.

Read the insight
06 · The 4D pairing

Two frameworks. One conversation.

The 4D Framework names four human competencies. The Capabilities and Limitations Framework names four machine properties. Each human "D" answers a specific machine property. Learn both, stop being surprised.

Read the insight

Want the boardroom version of all this?