Not a Cliff. A Staircase.

The panic is louder than the evidence. The history is more reassuring than the headlines.

This Has Happened Before

Every few decades, a new technology arrives and the same conversation starts.

The printing press will put scribes out of work. The assembly line will eliminate craftsmen. The spreadsheet will replace accountants. Computers will make whole industries obsolete.

Sometimes those predictions are partially right. The scribes did disappear. But the publishing industry that followed employed far more people than the scribes ever did. The pattern repeats. This isn't because technology is benign; it's because human beings are remarkably good at finding new things to do.

AI is the latest chapter in that story. And if you understand the pattern, it looks a lot less like a cliff and a lot more like a staircase.

The Assistant’s Transformation.

In the 1970s, a CEO would never touch a typewriter.

That was not laziness. It was simply how work was organized. Executives dictated. Assistants typed, printed, and mailed. The chain of production required multiple people and multiple steps to get a single document from idea to delivery.

Then personal computers arrived. Then email. The chain started to compress. Assistants still printed emails for executives to edit, then retyped the changes to send. That sounds absurd now but it was completely normal for over a decade.

Then something shifted. The technology improved enough, and became familiar enough, that executives started writing their own emails and company memos. The assistant's role did not disappear, it transformed. Travel coordination, event planning, stakeholder management, complex scheduling. Tasks that required judgment, relationships, and organizational knowledge: these are things no software could replicate.

The work changed. The people adapted.

Where AI Fits in That Pattern

AI, specifically Generative AI and AI Agents, is compressing another chain of production.

Tasks that previously required a specialist like drafting a first version, summarizing a document, writing basic code, analyzing data for patterns, can now be initiated by almost anyone. The specialist does not disappear. But the specialist who only does those tasks is at risk. The specialist who uses AI to do those tasks faster and then applies their judgment to what the AI cannot do. Context, nuance, and accountability make the human specialist significantly more valuable.

This is not a comfortable message for everyone. Some tasks will be automated. Some roles will change faster than the people in them can adapt. That is real and worth taking seriously. The macro pattern has held across every major technological shift in modern history: technology redefines tasks, work evolves, and new roles emerge. There is no evidence that AI is categorically different. There is a lot of noise suggesting it is.

Safe Harbor: Three Things You Can Do This Week

• List the tasks in your role that are repetitive and rule-based. These are the most likely to be automated. Knowing them is the first step to getting ahead of the change rather than behind it.
• List the tasks that require judgment, relationships, or accountability. These are your leverage. AI cannot be held responsible. It cannot build trust. It cannot read a room. Those capabilities are yours.
• Try using AI for one task you currently do manually. Not to replace yourself but to understand the tool. You cannot make good decisions about AI in your organization if you have never used it seriously.

Next week: We open the hood on AI agents to see how they think, decide, act, and where that process can go wrong.