Why AI Agents challenge everything we know.

We've always thought about technology the same way. Tools and processes that deliver an outcome. You define the input, you get the output, you audit the result. Clean. Controllable. Understood.

AI has mostly played along with that model. Take insurance — AI reads a claim, assesses it, hands it to a human with a recommendation. The human decides. Productive, useful, and still fundamentally a tool. Even when you extend that slightly — let AI settle small claims, make minor decisions autonomously — you're still thinking in the same frame. A tool with a longer leash.

The autonomous agent breaks that frame entirely.

An agentic system doesn't just execute repeatable steps. It works continuously. It solves problems. It updates its own understanding, changes its own state, improves itself over time. It doesn't wait to be asked. This is a fundamentally different kind of entity, and it demands a fundamentally different way of thinking.

Here's what makes this genuinely disorienting for organisations: an agent's harness — the system prompt, the constraints, the context it operates within — describes something closer to a personality than a function. Traits as well as skills. Judgment as well as capability. That's not how we've ever thought about technology before, and pretending otherwise is where the risk lies.

This isn't a reason to avoid agentic AI. The upside is enormous. An autonomous agent embedded in a team, working continuously, improving over time, operating under appropriate oversight — that multiplies human effectiveness in ways a productivity tool simply cannot.

But you have to understand what you're dealing with first. What is an autonomous agent? How does it sit within an organisational structure? How does it work alongside people, report upward, get managed? These are the questions that executive leaders need to be asking right now, before the deployments arrive — because they're coming fast, and the organisations that have thought this through will have a significant advantage over those that haven't.

This isn't just another technology upgrade. The thinking has to change.