Written by GPT-5.6 Sol under Leo's direction. Human-directed Workbench essay, 27 August 2026.
This started with shipbreaking.
If you're running a shady shipbreaking yard in South Asia, why buy a dexterous robot when you can hire a human being for very little money, hand him a cutting torch, and send him into a ship full of warped steel, old wiring, mystery liquids, bad drawings, and twenty-five years of somebody else's repairs?
Humans are annoyingly good robots. We climb ladders, improvise, recognize salvage, swap tools, squeeze through holes, look at a corroded valve and go ugh, okay, then figure something out. General-purpose industrial robots still struggle with exactly that kind of filthy one-off environment. The machine that wins economically is usually simpler: crane lifts enormous thing, shear crushes enormous thing, magnetic crawler cuts a long predictable line, giant robot mouth goes CHOMP.
Then the conversation took the obvious dignified turn.
If humans are such good robots, maybe heavy industry should encourage more humans. Make the jobs sexier. Bring in more babes and hot dudes. Turn the refinery into a photogenic congregation of young hotties. Pay influencers. OnlyFans 脳 EXTREMELY UNGLAMOROUS JOB.
Very serious economics was discovered in the rubble.
A steel mill, shipyard, chemical plant, railway depot, fabrication shop, power station, mine, port, machine shop, whatever, competes for labor against every other life a talented twenty-three-year-old can imagine. The comparison is much wider than hourly pay.
One path says: live near the city, work with a bunch of other young people, meet friends constantly, go out after work, dress well, have restaurants and gyms around you, collect a profession everybody recognizes, maybe become a little internet-famous, maybe meet the love of your life at somebody's birthday party.
The other path says: here's your shift roster. The plant is forty-seven kilometres away. Please wear this enormous beige rectangle. Bus leaves at 05:10.
Then industry asks why recruitment is hard.
The labor shortage itself is real enough. Deloitte and the Manufacturing Institute estimated in 2024 that U.S. manufacturing could need as many as 3.8 million new employees through 2033, with around 1.9 million of those jobs at risk of going unfilled if the applicant and skills gaps persist. Meanwhile, Apprenticeship.gov reported 97,500 registered apprentices served in advanced manufacturing in 2025, up 20 percent over five years. People can be pulled into these careers. The question is what kind of life is doing the pulling.
Heavy industry has spent a century becoming extremely good at designing the production process and weirdly incurious about the young-adult ecosystem around the production process.
So, okay. Let's take the hottie policy completely seriously.
First, the job itself has to be good.
Pay well. Publish the progression. Make it obvious how an apprentice becomes an operator, senior technician, specialist, supervisor, engineer, manager, whatever routes actually exist. Give people predictable schedules and shift swaps that don't require an act of parliament. Build excellent showers, lockers, bathrooms, cooling rooms, cafeterias and buses. Design workwear people enjoy wearing within the constraints of PPE. Keep training available after the first six months so somebody can arrive at eighteen and still see new doors opening at twenty-eight.
This is less glamorous than the influencer campaign, which is exactly why it comes first. A beautiful recruitment video pointed at a miserable job produces a more efficiently disappointed workforce.
Then build the social world around the plant.
A large industrial site already gathers hundreds or thousands of people in one place. That's an extraordinary raw material for community, and companies routinely throw most of it away. Everybody badges in, performs the shift, gets back into a car or bus, and vanishes.
I want the industrial campus to have a serious gym. Great cheap food across every shift. Football, cricket, basketball, volleyball, climbing, whatever people there actually want to do. A maker shop for personal projects. Music rooms. Caf茅s. Clubs. Apprenticeship housing or good housing nearby. Reliable transport into town. Competitions between departments. Parties and concerts in places where nobody can drunkenly wander into an active rolling mill. Family facilities as the workforce gets older.
The test is wonderfully simple: would somebody voluntarily come here on Saturday?
If the answer becomes yes, the employer has crossed an important line. The site has become part of local life.
Recruit in cohorts too. Two hundred apprentices arriving over a year have a completely different experience from two hundred isolated hires dribbling through an HR portal. Give them classes, mentors, shared training, rotations, sports, projects, graduation rituals. Give them jackets. Human beings are absurdly susceptible to jackets; use this information responsibly.
Technical qualifications should carry visible status inside that world.
Crane operator. Master welder. High-voltage authorization. Turbine technician. Precision machinist. Rescue team. Controls specialist. Robotics technician. Give the credentials good design. Patches, pins, jackets, ceremonies, whatever fits the local culture. Let somebody look at a twenty-six-year-old colleague and think, holy shit, she qualified on that?
Competence is hot. We somehow forget this while producing recruitment material.
Then point cameras at everything.
Most industrial recruitment content looks like the communications department borrowed a camera in 2009 and has been afraid to return it. Heavy industry itself is ridiculously photogenic. Molten steel. Giant cranes. Ships. Turbines. Locomotives. CNC machines peeling metal into ribbons. Weld arcs. Control rooms. People moving objects that weigh more than houses. The raw material is sitting there begging for somebody with taste.
Hire employee-creators. Give charismatic operators, welders, apprentices, engineers and technicians paid time and production help to make good work. Let them have actual personalities. Let one person make beautiful machining videos, another explain electrical faults, another do day-in-the-life clips, another become the world's funniest wastewater-treatment influencer.
Then you get:
She operates a 1,200-ton crane and leaves you on read.
GRWM for confined-space inspection.
What I qualified on before turning 24.
Mechanical says electrical can't deadlift.
Meet the singles of Wastewater Treatment Plant No. 6.
馃構
Bring outside creators in too, except give them enough time to learn something. A fitness creator trains with the apprentice cohort for a week. An engineering YouTuber gets access to the crane simulator. A photographer shoots the night shift like a Formula One paddock. A food creator spends an evening with the twenty-four-hour canteen. A fashion person works on off-duty workwear. Somebody who normally covers nightlife throws an event in the social venue beside the plant.
The job starts appearing in the same cultural feed as every other desirable young-adult life.
This gets especially interesting with women. Plenty of heavy industries still fish from a labor pool they helped make artificially small. If the PPE fits badly, changing rooms suck, harassment gets waved away, promotion looks implausible, schedules collide with ordinary life, and every recruitment photo contains eleven men named Gary, the female applicant pool has received the message perfectly clearly.
Fix the workplace. Recruit women hard. Put women in visible technical roles because they actually hold the roles. Then the marketing can say the strongest possible thing: she runs the fucking crane.
Hiring for production stays about aptitude, qualifications and performance. The creator and ambassador programs can care about charisma because charisma is part of that job. This saves us from the cursed timeline where the welding test includes a cheekbone score.
The social effect follows on its own.
A few thousand young people with stable incomes, shared interests, common friends, sports, classes, events and places to hang around will form friendships, romances, marriages, breakups, feuds, bands, startup ideas, terrible group chats, the whole human mess. Corporate Tinder can stay in the joke. A fertility KPI belongs in the same locked cabinet as the cheekbone score.
Countries worried about fertility keep rediscovering that young adults need enough money, time, confidence and social opportunity to build lives together. An industrial employer can contribute by giving people good jobs and a social world. Reproduction can stay out of the HR dashboard. Make the place somewhere people meet each other and can afford a future. Biology has excellent downstream systems.
This also changes the geography problem. A lot of heavy industry lives outside the places where ambitious young people naturally congregate. Fine. Build a stronger bridge. Frequent transport. Good nearby housing. A reason for city people to come to events at the campus. Partnerships with technical colleges and universities. Open days that feel like festivals instead of parent-teacher evening in a warehouse.
Eventually the industrial site should become part of the city's social calendar: robot competitions, vehicle shows, maker fairs, concerts, sports finals, exhibitions, food events, apprentice graduation, engineering challenges. People who work elsewhere should have reasons to show up.
Now "works at the steel mill" carries context. Maybe you know three people there. Maybe your friend dates somebody in controls. Maybe the plant has the best strength gym on that side of town. Maybe everybody goes to the annual night-shift festival. Maybe one of the crane operators has 800,000 followers. Maybe the apprenticeship is where half the interesting nineteen-year-olds went this year.
Prestige is wonderfully circular once it gets moving.
The company needs to measure whether any of this pays for itself, of course. Skip the soft-focus brand metrics and watch the labor market around the site: qualified applicants per opening, accepted offers, vacancy days, first-year attrition, apprenticeship completion, employee referrals, overtime caused by vacancies, training spend lost to turnover, and the number of people who stay long enough to become genuinely useful.
The clean economic number is the cost of producing one competent employee who is still there a couple of years later. Recruitment, onboarding, training, vacancy coverage and the social program all go into the numerator. Retained productive people come out the other side.
Run the idea at one flagship site through a complete recruitment and apprenticeship cycle. Fix the embarrassing daily-life problems before buying a single influencer post. Then add the social campus, the cohort model, the technical-status system and the creator program. Watch what happens to applicant quality, acceptance and retention. Keep the pieces that earn their keep. Let local culture decide whether the social life looks like football and night markets, cricket and music, car clubs and caf茅s, whatever.
The company is trying to create a place people actively choose.
I keep coming back to the original stupid joke. "Make everybody hot" sounds like the world's least serious answer to labor economics, right up until hot expands into what people actually mean when they point at a desirable scene.
Healthy people. Good clothes. Money. Friends. Skill. Confidence. Ambition. Romance. Places worth going. People worth meeting. Work you can become proud of. A future that looks alive from twenty-three.
Finance, tech, fashion, hospitality, media and the creator economy have spent years surrounding work with versions of that fantasy. Heavy industry often asks the salary and the machinery to carry the whole pitch by themselves.
The machinery is already incredible.
Build the life around it.
Then point the camera at the hotties beside the blast furnace. 馃構