We were all farmers at some point. Then we mastered the art of building tools, and most of us forgot how to work the ground. Nobody calls that a tragedy.
The same conversation is happening with AI right now, except it has been dressed up as an existential threat. Junior people will lose entry-level skills. Spreadsheets will write themselves. Contracts will read themselves. The next generation will become non-thinking zombies.
Or maybe the skills just evolve
Maybe. Or maybe entry-level skills are about to evolve, and the people doing them will get to use their brains for something more interesting. That is not a naive hope; it is the pattern of every tool transition we have been through. We did not lose our minds when we stopped ploughing by hand — we redirected them. We did not stop thinking when calculators arrived, or spreadsheets, or search engines. Each time, the mechanical part of a job was handed to a machine, and the human part moved up a level.
The risk isn’t the tool. It’s opting out.
The genuine risk is not that AI erodes skills. It is treating the transition as something that happens to you rather than something you shape. The junior analyst who uses AI to clear the mechanical work and spends the freed hours learning judgment will outrun the one who either refuses to touch it or outsources their thinking to it wholesale. Entry level is not disappearing. It is being redefined, and the people who engage with that deliberately get the better end of it.
This is also how we think about building AI for other businesses: take the low-value work off people so they can spend their day on the part only they can do, not hollow out the role. It pairs with a point I have made elsewhere about the human judgment no model can reach.
The doom framing makes a good headline. History makes a better guide. Where do you think the line really falls: skill loss, or skill evolution?