Everyone Thinks “AI Jobs” Means Programmers

Ask a room full of parents, teachers, or career counselors what an “AI job” looks like, and you will get a nearly unanimous answer: someone, somewhere, writing code. Maybe a computer science degree. Maybe a coding bootcamp. Definitely a laptop covered in stickers. This is the story everyone is telling right now — in commencement speeches, in district strategic plans, in the slightly panicked group chats of parents wondering whether to push their kid toward Python in seventh grade.

It is also, according to the people actually doing the hiring, a fairly outdated story.

The current narrative around AI and careers tends to collapse into one of two headlines: either AI is going to automate away every entry-level job in existence, or AI is creating a gold rush for anyone who can write a few lines of code. Both narratives share a hidden assumption — that the AI economy has basically one entry point, and it runs through software engineering. That assumption is doing real damage to how schools counsel students, how career-changers plan their next move, and how the rest of us think about whether we are “technical enough” to have a future that involves AI at all.

Here is the much stranger, more interesting truth this post is built around: the fastest-growing parts of the AI economy right now are not primarily about building models. They are about governing them, auditing them, explaining them to regulators, deciding where they should and should not be deployed, and managing the human fallout when they are. None of that requires a computer science degree. Almost none of it existed as a job description five years ago. And surprisingly little of the conversation about “careers of the future” mentions it at all — which is exactly why this post exists.

What’s Actually Happening: A Job Market Being Rebuilt in Real Time

Start with the scale of it. The World Economic Forum’s Future of Jobs Report 2025, which surveyed more than 1,000 companies representing over 14 million workers worldwide, estimates that by 2030 roughly 170 million new jobs will be created globally while 92 million are displaced — a net gain of about 78 million jobs, even after accounting for the disruption (World Economic Forum, 2025). Technology-related roles, including AI and machine learning specialists, were named among the fastest-growing job categories in percentage terms (World Economic Forum, 2025).

That headline number is the one everyone quotes. The more useful number, for our purposes, is what kind of roles are actually filling that growth. A 2025 analysis from the AI Workforce Consortium — a coalition that includes Cisco, Accenture, Google, IBM, Microsoft, and SAP, among others, examining 50 technology and specialized-support roles across G7 countries — found that seven of the ten fastest-growing technology roles are AI-related, and that two of the standout categories were AI Risk & Governance Specialist and roles built around AI ethics (Cisco, 2025). Demand for AI governance skills specifically grew by roughly 150% over the year studied, and demand for AI ethics skills grew by about 125% (Cisco, 2025). Demand for skills tied to responsible AI deployment — the work of making sure systems are used appropriately once they are live — grew even faster, at roughly 256% (Cisco, 2025).

The most visible symbol of this shift sits at the top of the org chart. Two years ago, the title “Chief AI Officer” barely existed outside of a handful of tech companies. According to research from the IBM Institute for Business Value released in 2026, 76% of surveyed organizations reported having a Chief AI Officer that year, up from just 26% the year before — and not only at technology companies, but at firms like Heineken, Nike, and CVS Health (IBM Institute for Business Value, 2026). The role itself has matured alongside the statistic. As Jacob Dencik, a research director at the IBM Institute for Business Value, put it: it used to be that chief AI officers were mostly figureheads acting as AI evangelists; the job has since become considerably more operational (IBM Institute for Business Value, 2026).

It is not only the executive suite. Cisco’s Chief People, Policy & Purpose Officer, Francine Katsoudas, summarized the broader pattern this way when the Consortium’s findings were released: “As AI reshapes our world and work, people remain at the center” (Cisco, 2025). That framing matters, because it cuts against the assumption that the growth in AI-related work is happening to machines instead of around them. Stanford economist Erik Brynjolfsson, director of the Stanford Digital Economy Lab, has gone further, predicting that entirely new role categories — he has floated examples like a “chief question officer” and an “agent fleet manager” — will emerge as organizations need people who can frame the right problems for AI systems and supervise growing teams of AI agents, alongside a wider class of “citizen developers” who can build useful software without deep coding expertise (Brynjolfsson, as cited in National Today, 2026).

Put plainly: the people who can frame a useful problem for an AI system to work on, and who can tell when its answer is wrong, are becoming more valuable, not less — and that skill has very little to do with whether you can write a for-loop.

Where These Careers Are Already Showing Up

None of this is hypothetical. It is already changing what career exploration, course selection, and job searching look like at every stage — even for people nowhere near the job market yet.

For Elementary and Middle School: Planting Wider Seeds

Nobody is expecting a sixth grader to know what an AI governance specialist does, and they shouldn’t have to. What matters at this stage is exposure that isn’t limited to “coder.” Career-day rosters, classroom read-alouds about future jobs, and “day in the life” activities can just as easily feature someone who writes the rules for how a hospital’s AI scheduling tool gets used, or someone who checks whether a school’s AI tutoring software treats every student fairly. The goal is not career selection. It is widening the picture of what “working with AI” can mean before the narrower version calcifies.

For High School: Pathways Beyond the Computer Science Track

This is where the stakes get real. Students choosing electives, dual-enrollment courses, or early certifications are often funneled toward a single AI-adjacent pathway: computer science. But the fastest-growing AI roles described above draw just as heavily on law, communication, psychology, statistics, and ethics coursework as they do on programming. A student with strong writing and argumentation skills and an interest in policy is a more natural fit for AI governance work than a student who only knows Python. Counselors and teachers can start naming that explicitly — not as a consolation prize for students who “aren’t into coding,” but as a legitimate, well-paid pathway in its own right.

For Recent Graduates and Early-Career Professionals: The Jobs Are Real, and They’re Hiring Now

For anyone already in the job market, this is the most actionable section. AI governance, AI ethics, AI adoption, and AI project management roles are being posted right now, often by companies that are not traditional “tech” employers at all — banks, hospital systems, retailers, manufacturers. Mid-level AI governance roles have been reported in the $130,000–$180,000 range, with senior or specialized positions commanding more, particularly for candidates who pair legal, compliance, or policy backgrounds with technical AI literacy (HeroHunt.ai, 2026). A communications degree plus a genuine understanding of how AI systems make decisions can be a stronger application for an “AI adoption manager” posting than a computer science degree with no organizational experience at all.

Risks and Tradeoffs: What Could Go Wrong

None of this is a reason to relax. The same forces creating these new careers are creating real friction, and a balanced look at the AI workforce has to sit with three of them honestly.

Workforce disruption is uneven, and entry-level workers are absorbing more of it. Research from the Stanford Digital Economy Lab, led by Erik Brynjolfsson alongside colleagues including Bharat Chandar, has documented a measurable and disproportionate effect on entry-level employment in fields like software engineering and customer service, concentrated among workers in their early twenties (Fortune, 2026, citing Brynjolfsson et al.). New senior-sounding titles at the top of an org chart do not automatically translate into more first jobs at the bottom of it — and PricewaterhouseCoopers’ newly released 2026 Global AI Jobs Barometer found that AI-exposed junior roles are now seven times more likely than less-exposed junior roles to demand traditionally senior skills like leadership and strategic thinking, even as overall early-career postings in highly AI-exposed sectors have flattened (PricewaterhouseCoopers, 2026).

The training pipeline has not caught up with the hiring need. The same AI Workforce Consortium report that documented soaring demand for governance and ethics skills also found a critical shortage of workers who actually have them, alongside generative AI, large language model, and AI security expertise — serious enough that Consortium members collectively committed to upskilling 95 million people worldwide over the next decade (Cisco, 2025). A fast-growing job category is not much comfort to a school or a career-changer if nobody nearby is teaching the skills it requires yet.

Credential inflation is real, and it is accelerating. As AI literacy becomes a baseline expectation, the market for AI certifications has exploded: from fewer than a dozen recognized credentials in 2023 to more than 100 by 2026, spanning vendor platforms, foundational literacy, technical machine learning, domain-specific, and ethics-and-governance categories (TrainAI, 2026). That growth has produced what some hiring analysts now call “certification fatigue” — resumes listing “AI proficiency” have become so common that the claim alone no longer differentiates a candidate, and employers increasingly weight demonstrated, applied experience over the certificate itself (TrainAI, 2026).

What Educators, Parents, and Mentors Can Do Now

Start by retiring the single-pathway story. When career-day rosters, “in-demand jobs” posters, and college-and-career nights only feature software engineers and data scientists, they are quietly telling every student who is good with people, words, or rules that the AI economy was not built for them. It was. Update those lists.

Invite people from the new orbit, not just the old one. An AI ethics consultant, a compliance officer who works on AI risk, or a project manager who led an AI rollout at a local hospital or bank can be a far more useful guest speaker right now than another software engineer — precisely because almost no one is inviting them yet.

Teach the vocabulary early. Students who can correctly use and explain terms like AI governance, AI ethics, and AI adoption — even at a basic level — walk into internship and job interviews with a real advantage, because most of their competition still cannot.

And resist outsourcing career guidance entirely to a list of “AI-proof jobs.” The far more useful skill is teaching students to evaluate any job, including ones that do not exist yet, by asking what human judgment it actually requires.

What Leaders Should Be Considering

For school and district leaders, the open question is whether AI literacy programming lives only inside the computer science department or gets threaded through career and technical education, business, law-and-government, and humanities pathways as well. The data above suggests the latter is where the job growth actually is.

For business and organizational leaders — and this post is for you too — the open question is whether your organization is staffing AI governance and adoption work as an afterthought bolted onto IT, or as a genuine cross-functional function with a seat at the table. The IBM research on Chief AI Officers suggests the organizations seeing the strongest return on their AI investment are the ones that already made that call (IBM Institute for Business Value, 2026).

For workforce-development professionals and local employers, there is a concrete near-term opportunity: partner with schools and community colleges now, while these pathways are still forming, rather than waiting until the talent pipeline problem documented above becomes a crisis.

A Forward-Looking Close: Preparing for Jobs With No Name Yet

This is, in the end, not really a story about technology. It is a story about how badly we want certainty before we are willing to prepare for something. Deloitte CEO Cathy Engelbert, citing World Economic Forum research, has put a number on exactly that discomfort: “65 percent of them will eventually have a job that doesn’t exist today” — referring to today’s school-age children (Engelbert, as cited in MSU Broad College of Business, 2019).

That statistic predates the current AI moment by several years, and yet it has aged into something closer to a description of right now than a prediction. If the fastest-growing career paths are ones that did not have names five years ago, what exactly are we promising someone when we tell them to “pick a career”? Maybe the more honest promise — for a student, for a career-changer, for anyone reading this who is wondering whether they are already behind — is to pick a problem worth getting good at solving, and trust that the title attached to it will eventually catch up.

Next week, in the final post of this series, we turn from the careers themselves to the daily reality of working alongside AI once you are in one of them: what it actually looks like to be an “AI-augmented professional,” the literacy that requires, and where human verification still has to be the last word.