This is part 2. Part 1: Weak AI, Strong Fear was about the fear that AI will wake up and go off on its own. In short: it won’t. Today’s AI is weak, a tool without goals of its own.
But even weak AI changes how we work. And the first to feel it are the people who spent years right at the cutting edge.
First Scrum, then Kanban, then spectating
Hardly any profession has had to reinvent itself as often in recent years as developers. Always knowing the latest software, learning the next framework, joining the next project management method: first Scrum, then Kanban, then something with “Agile at Scale”. If you weren’t up to date, you were out.
And today? Today many of them sit in front of their screens watching Claude Code turn the build checkmark green.
Quick translation, because that was jargon: Claude Code is an AI tool that writes code on its own. The “build” is the moment the code gets assembled into a finished program. When the checkmark turns green, everything worked. That used to be a human’s job. Now the human watches.
The meme above sums it up, and it’s one of the reasons I’ve reoriented myself. I don’t want to be the one who just watches. I want to be the one who decides what the machine should do.
What the machine replaces, and what it doesn’t
You can draw a conclusion from this that’s less bleak than it sounds. If a machine writes code like a senior developer, weak AI is replacing a fairly repetitive task. It isn’t replacing the person. Someone still needs to know what should be built, for whom and why.
You see this most clearly in art and culture. In October 2026, China released “Sanxingdui: Future Past”, the first fully AI-generated feature film. On a budget of around 1.5 million dollars, it took in only about 180,000 dollars in its first week. Audiences stayed away.
That doesn’t surprise me. Whoever makes a film usually has a bubble: people who care about exactly this story, retell it and multiply it. No matter how small it is at first. A bubble like that grows around people, not around computing power.
A friend recently told me that at some point studios will just buy the rights to actors’ AI avatars, and the actors can skip acting altogether. Maybe that’s technically possible. Whether anyone will go to the cinema for it is another question. The numbers from China currently suggest otherwise.
Where does all that talent go?
Which leaves the question that fascinates me most right now: where do developers go when their work is suddenly needed less? And probably paid less, too.
I think the answer is more exciting than it sounds. Until now, the really good technical people sat almost exclusively where there was money for expensive technology: corporations, agencies, venture-backed start-ups. A sports club, a food bank or the corner shop could never afford someone like that.
If expensive technology now becomes a soft skill that AI makes available to everyone, that shifts. Highly skilled people end up in fields that simply had no budget for them before. On balance, that’s a net gain, everywhere.
In theory, a small corner shop’s communication could then be just as good as that of a big discount chain. But the greatest potential I see is in social organisations, clubs and associations, and everything that makes living together simpler and better at a hyperlocal, regional level. Exactly where people work for people, and where until now every euro spent on technology was one too many.
Why I’m still optimistic
I look at the future with optimism. Because, as I said: AI is weak, not strong. It’s a tool, and we can use it as one.
And it democratises knowledge. Anyone who wanted marketing used to need an agency. Today a small agent is often enough to do your own marketing. You don’t need a masterclass for that. Just talk to your agent, it’ll tell you what it needs from you.
And if you ever get sick of it: switch the language model. I’ve worked with ChatGPT, Google’s Gemini, Qwen, Claude, Mistral and Llama, and each has different strengths. Which one suits your business depends on what you do. That’s exactly what I’m happy to find out for you.
Source on the AI film: Fortune: China’s first AI film flops at the box office