Young Workers Are Ditching Screens for Sawdust and Copper Wire

Key points

• Entry-level white-collar jobs in accounting, law and software are increasingly being replaced by AI automation tools.
• Young workers are turning to heritage crafts such as bookbinding, metalwork and boatbuilding as alternatives.
• University degrees, long seen as a reliable path to stable work, are losing their edge in an AI-disrupted market.
• Creative fields including design, writing and visual art now compete directly with AI tools that produce finished work in seconds.

For a long time, the deal seemed simple: study hard, get a degree, land a desk job. That deal is fraying.

A university qualification still carries weight, but the entry-level roles it used to unlock are shrinking fast. AI automation tools, software that handles tasks once done by junior employees, are doing the grunt work in accountancy firms, law offices and tech companies at a fraction of the cost. The first rung of the career ladder is getting harder to find.

What jobs are actually under pressure?

The clearest squeeze is happening in offices. Budding accountants are competing with software that can process thousands of invoices overnight. Trainee lawyers find AI can draft a contract in minutes. Junior software engineers, once indispensable, now share the floor with AI code-writing tools that never sleep and never ask for a pay rise.

Creative careers are feeling it too. A graphic designer starting out in 2025 is not just up against other designers. Tools built on large language models, the same technology behind chatbots like ChatGPT and Claude, can now produce polished artwork, detailed copy and full design briefs from a short typed instruction. That used to take hours and cost real money.

So where are young people going instead?

Increasingly, towards things a machine cannot easily do with its hands.

The Guardian reported a noticeable drift toward heritage crafts: bookbinding, copper and ironwork, traditional boatbuilding. These are trades that require physical skill built over years, judgment that comes from touch and experience, and a final product that carries the mark of a human maker. None of those qualities are easy to automate.

Picture a 24-year-old who spent two years learning to code, watched AI tools absorb the beginner tasks, and now spends her days in a bookbinding workshop stitching pages by hand. The work is slower, the tools are older, but nobody is releasing an AI update that threatens her Thursday afternoon.

There is something practical underneath what might look like nostalgia. A well-made piece of furniture, a hand-riveted copper bowl, a traditionally built wooden boat: these things fetch real money precisely because they are rare, and because the people who want them understand a machine did not make them.

What should young people actually think about this?

No single career path is completely safe, but the risk is not equal across every job. Roles that deal in physical materials, in-person care, bespoke craftsmanship and deep human judgment are holding up better than paper-shuffling done at a desk.

That does not mean scrapping every plan and buying a lathe. It does mean thinking honestly about which parts of any job a cheap software tool could do in five years, and building skills around the parts it cannot.

The tools will keep improving. The people who combine a craft skill with even a basic understanding of the technology around them will probably land in the strongest position of all.

Common questions

Are all office jobs disappearing?

Will learning a trade actually pay?

Not all of them. AI is eroding specific tasks inside jobs, particularly repetitive, rules-based work done by people early in their careers. Senior roles involving client relationships, complex decisions and accountability are much less exposed for now.

Increasingly, yes. Skilled trades already face a shortage of trained practitioners in many countries, and handmade or bespoke goods command premium prices in a market flooded with identical machine-made products.