Gen Z Is Steering Away From Code, and Tech Will Feel It in 2030

Fortune reports that Gen Z is steering toward healthcare because it reads as AI-proof. The same survey data ranks chiropractors, doctors and paramedics among the unhappiest workers in the economy. So young people are choosing work they may well dislike, because they think it will still exist.

Picture the eighteen-year-old in a community college advising office picking a paramedic program over a computer science degree. Then add the thirty-two-year-old engineer in Bangalore who left a fourteen-year IT career to farm mushrooms. Neither shows up in a layoff count. We think they matter more than this year's headcount numbers do.

This is a supply story, and supply stories move slowly until suddenly they don't. Enrollment decisions made in 2026 turn into hiring pools in 2030. If the cohort entering college now routes around software in real numbers, the industry that spent 2026 congratulating itself on doing more with fewer people will spend 2031 finding out what a genuine shortage of trained mid-career engineers costs.

A license is the protection they're after

What Gen Z seems to be selecting for is legal protection. Forbes published a list of fifteen fast-growing jobs AI can't touch that pay up to $175,000, and most of them share a license, a board or a statutory scope of practice. Nurse anesthetists, physician assistants, electricians and speech-language pathologists aren't safe because their tasks are impossible to imitate. They're safe because a regulator decides who may legally do the work.

Software has nothing like that. Anyone can call themselves a developer, and in 2026 that's exactly the problem. The Korea Times documents young Korean workers taking the sharpest losses from automation, and the AI Footprint newsletter tracks the same hiring friction for young workers in the United States. When the entry tier of an unlicensed profession collapses, the rational move for a teenager is a field where an institution rations entry, instead of a hiring manager's mood and a headcount freeze.

Israel is the counterexample we're watching. Calcalist frames that country's tech layoffs as a chance to redistribute talent, on the theory that displaced senior engineers will seed defense, health and industrial firms that could never outbid the big platforms before. That can work when the talent already exists. It does nothing about a cohort that never entered.

Hiring signals stopped telling employers much

The Wall Street Journal reports job seekers racing to AI-proof their résumés. Business Model Analyst found job ads paying a 62 percent premium for AI skills and pointed out the obvious catch: typing the word costs nothing. Once a premium attaches to a keyword, everybody adds the keyword, and it stops meaning anything.

The Journal's reporting on North Korean operatives faking their way into US companies is the extreme version. A fully remote hiring funnel built on self-reported credentials, video calls and take-home tests can be beaten at scale by a determined adversary. Every fraud control companies add in response raises the cost of being an unknown candidate that nobody can vouch for.

That ties together two headlines that look unrelated. Axios reports Gen Z wanting more office time, and Toptal's data shows experienced tech professionals seeing stronger demand despite the layoff wave. We read both as verification. Showing up in person and having a long track record someone can check are the two cheapest ways left to prove you are who you say you are. Named-employer tenure and referrals gain value in that world, which tilts things toward incumbents and away from newcomers. Portfolios and open-source work still count for something, though that's getting contested as AI-assisted output becomes hard to tell apart.

Colorado's AI rules hint at a credential layer for tech

Colorado's AI rules, tracked this week in the AI Footprint roundup next to grid interconnection queues, create a class of work that has to be done by an identifiable, accountable person: impact assessments, documented human review, disclosure obligations, bias auditing. No Jitter's rundown of six AI ownership problems in the workplace describes the same vacuum from inside companies, where nobody can say who owns a model's output when it goes wrong.

Regulation is how unlicensed professions become licensed ones. Accounting, engineering and financial advice all went that way after failures made accountability a legal necessity. If the next five years bring a serious AI liability event, the compliance-adjacent tech jobs that look like overhead in 2026 could become the credentialed tier of the industry. That's a big if, and we're guessing on timing.

There's a macro clock running too. Investing.com expects a Fed pivot after Friday's labor data, and pending home sales are falling in Seattle and other tech-heavy metros as employment softens. Cheaper capital would restart hiring. It would restart it into a pipeline that's already rerouting, and that's how you get wage spikes instead of broad rehiring.

Seniors get bid up while juniors can't get in

The pain inside tech isn't spread evenly. Toptal's report shows senior professionals in stronger demand even as 2026 cuts pass all of 2025, with Monday.com joining roughly twenty companies that pinned reductions on AI, and Patreon and Lucid cutting 20 and 18 percent respectively. Experience is being bid up while entry is rationed. The Conversation's research on why AI training can backfire for older workers adds pressure from the other end: badly designed upskilling can signal deficiency instead of building capability, and it nudges experienced people toward the exit. That leaves juniors who can't get in, seniors who are valued but selectively alienated, and a middle that keeps thinning.

Specific expertise still clears the market. Research from the recruiting firm South found that DIY hiring for niche stacks like Shopify development keeps failing, and 2K's launch of Small Axe Studios shows companies still assembling specialized teams from scratch. Generalist availability doesn't clear it.

Our best guess for 2030

Here's the scenario we think is plausible. Assume CS and bootcamp enrollment falls meaningfully through 2028 as this cohort reroutes. Assume AI tools keep compressing junior task volume while leaving system design, integration and accountability work intact. Assume at least two more states follow Colorado with AI governance rules.

Run those forward and by roughly 2030 the market is tight at the top and empty at the bottom. Companies that cut their junior tier in 2025 and 2026 have no internal supply of engineers with five years of experience, and no outside supply either, because the people who would have become those engineers went to nursing school. Pay for verified mid-career engineers climbs sharply. Employer-funded apprenticeships come back out of necessity rather than goodwill.

Any of those assumptions could be wrong, and enrollment could hold up better than we expect.

If you're early in your career, the licensure logic Gen Z is following isn't crazy, but leaving tech entirely looks like an overcorrection given what a 2030 shortage would mean. And if you do pivot into a licensed field, check the job satisfaction data first. Safe and happy aren't the same thing.

If you're mid-career, build the accountability and governance experience that regulation is about to make scarce. BCG's North America chief pointed to judgment as the skill that matters most in the AI era, and compliance-heavy work is where judgment gets written down. Put your name on shipped outcomes a stranger could verify without taking your word for it, because the 62 percent AI premium goes to demonstrated deployment and not to the phrase on a résumé. Selective office presence helps here too, as a way of being known.

If you manage a team, the contrarian move is to restart junior hiring now, while the pool is deep and cheap.

Two things we'll be watching: whether CS enrollment actually drops over the next couple of intake cycles, and whether any state beyond Colorado passes AI rules that require a named, accountable human. If both happen, the 2030 shortage stops being a guess.