Coding Alone Isn't Enough Anymore

Coding alone isn't enough anymore. And the data is stronger than most people realize. A March 2026 Resume.org survey of 991 U.S. hiring managers found that 57% now rate creative thinkers as more valuable than coders in the age of AI. This isn't anti-technical — it's a repricing of which human skills still produce scarce, non-automatable value. And agentic AI is accelerating it.

What the hiring data shows

The Resume.org numbers are worth sitting with. In a survey fielded in March 2026 across 991 U.S. hiring managers, 57% said creative thinkers bring more value than coders as AI automates more technical work. 81% said creative skills are becoming harder to find in candidates — not because fewer people are creative by nature, but because most pipelines haven't been designed to develop, signal, or reward creative judgment at scale. And a striking 14% said they had already eliminated coding-focused roles in favor of positions requiring more creative problem-solving.

Dig a level deeper and the picture sharpens. 76% of managers cited creative work as genuinely hard to replicate with AI — naming things like originality, narrative judgment, cross-domain synthesis, and taste. 72% said creative thinking now drives their best strategy and decision-making work, meaning it shows up in the rooms where resources are allocated and priorities are set. Interestingly, the sectors leaning hardest into this shift are the ones you might expect to lean the other way: financial services (66%) and technology (65%) reported the highest premium on creative skills — industries that already have the most direct exposure to AI-driven automation and are therefore most attuned to what remains scarce.

One reading of this is that AI hasn't made technical skills less important — it has made pure technical skills less differentiating. When models can generate working code on demand, the premium shifts to whoever can decide what to build, why it matters, and how it fits into a broader system of people, incentives, and outcomes. That's a creative question, not a syntax question.

The agentic shift changes the question

The hiring data reflects a tectonic shift happening underneath it: the move from AI as a tool you prompt to AI as an agent that takes goals, makes decisions, and executes tasks — sometimes across multiple systems, sometimes for hours at a stretch, sometimes without a human in every step. The numbers on this transition are remarkable.

Deloitte's State of Generative AI in the Enterprise 2026 report found that 74% of enterprises plan to deploy agentic AI within the next two years — up from roughly 23% today. Read that again. In a 24-month window, enterprise agentic AI deployment is projected to roughly triple. This isn't a curiosity. It's the largest single forward commitment to a new software category most leaders have seen in a decade.

Gartner projects that 40% of enterprise applications will have task-specific AI agents built in by the end of 2026, up from less than 5% in 2025. By 2028, they forecast 15% of day-to-day work decisions will be made autonomously by AI agents. That number sounds small until you realize it includes decisions about scheduling, routing, approvals, procurement, ticket triage, and countless other moments where a human used to weigh in and now won't.

Recruiting is moving right alongside it. Korn Ferry's Talent Acquisition Trends 2026 report found that 52% of talent acquisition leaders are actively adding AI agents to their teams — not as assistants, but as team members with responsibilities. And when those same leaders were asked what skill matters most in a world of AI agents, 73% named critical thinking. Not prompt engineering. Not coding. Critical thinking.

Agents take goals, make decisions, and execute — without a human in every step. That reframes the question from "what will AI do to my job?" to "what am I actually for?"

This is the real unlock. When agents are doing the middle work — the drafting, the routing, the first-pass analysis, the boilerplate code — the scarce thing is whoever can define the goal well, question the framing, notice when the output is subtly wrong, and decide what to do next. Those are creative and critical-thinking capacities, not technical ones.

What the research confirms

Three independent bodies of research point in the same direction as the hiring data.

The labor-market picture: net new work

The World Economic Forum's Future of Jobs Report 2025 — built from employer survey data representing more than 14 million workers across 22 industries — projects 170 million new roles created against 92 million displaced by 2030, a net addition of 78 million jobs globally. The headline isn't "AI eats everything." The headline is that the composition of work is being rewritten, and the skills the WEF ranks fastest-rising are cognitive ones: analytical thinking, creative thinking, resilience, curiosity, and technological literacy used together — not in isolation.

The psychology of the response: "Poets Over Quants"

A striking 2026 paper in the Personality and Social Psychology Bulletin out of Kellogg — nicknamed "Poets Over Quants" — ran nine studies with 2,320 participants examining how people adjust their skill investments when they perceive AI as a threat. The finding: perceived AI threat reliably drives people toward creative skills, and the effect held even when participants were shown evidence that AI can produce creative outputs. In other words, the pull toward creativity isn't just a rational response to the current capability frontier — it's a deeper intuition that creative judgment is where distinctly human value lives. Whether that intuition is correct or not, it is shaping education, career choices, and hiring behavior right now.

The economics: complement, not substitute

Perhaps the most important empirical study of the moment comes from Oxford's Internet Institute. Stephany and Teutloff (2024) analyzed 10 million UK job vacancies and found that the complementary effect of AI is up to 50% larger than the substitution effect across most occupations — and that workers with AI skills command a 23% wage premium. The people winning in this labor market aren't competing with AI; they're directing it. The premium is on judgment and integration, not on replication of what the model already does.

Three angles: higher ed, K-12, parents

If you believe the data, the question becomes: what changes for the institutions that develop talent? I think about this from three seats.

Higher ed

The most valuable graduates of the next decade will know what to build and why — not just how. They will be fluent enough with agents to direct them, supervise them, and catch their failures, but their differentiator will be framing problems that are worth solving in the first place. That means programs that pair technical literacy with the humanities, the social sciences, design, ethics, and domain depth. Pure syntax training produces graduates whose work is the most automatable part of the stack. Training in judgment, argument, synthesis, and stewardship produces graduates whose work sits above the automation line.

K-12

If an agent can already do a task at age-level quality, the teaching job is to focus on what comes after the task. Reading, writing, and math remain non-negotiable foundations. On top of them, creativity, curiosity, and critical thinking have to be taught as first-class subjects — with real rigor and real assessment, not as slogans. Kids who can only execute recipes that AI executes faster will be out-competed by kids who can question the recipe, improve the recipe, and explain why they chose it.

Parents

I am a parent too, and I watch for a simple signal. If my kids are being taught to memorize what AI already does, I worry. If they are learning to ask better questions, create original work, and judge the quality of outputs — including AI outputs — they're in the right place. That's the rubric I apply to every curriculum decision, every extracurricular, and every conversation about homework. We are raising a generation that will work alongside agents for their entire careers. The earliest habits matter most.

What this means for our role

At Our Community Tech, we build for nonprofits and the communities they serve. That puts us in a specific seat — close to mission-driven leaders, close to students, close to families navigating the tradeoffs of technology in real time. The agents are here. The capability curve is steep and getting steeper. Our role, I believe, is not to evangelize and not to warn from the sidelines — it's to be thoughtful stewards.

Stewardship means helping the organizations we serve adopt these tools in ways that make their teams stronger, not thinner. It means building things that expand human judgment rather than replace it. It means being honest about what agents are good at (scale, consistency, draft-zero work) and where they still fail (context, ethics, nuance, taste). And it means remembering that technology, for all its power, is only ever half the story.

Technology alone is not enough. It's technology married with the liberal arts, married with the humanities, that yields us the results that make our hearts sing.

— Steve Jobs, as quoted by Tim Cook at the 2017 MIT Commencement

Cook delivered that line at MIT's 2017 commencement in the context of a different technology wave, but the framing travels. The companies, schools, and nonprofits that will matter most in this decade won't be the ones that deployed the most agents. They'll be the ones that figured out how to pair powerful tools with durable human judgment — the creative thinking, the critical thinking, the moral thinking — and did both deliberately.

Coding alone isn't enough anymore. It hasn't been for a while, honestly. The data is finally loud enough that we can stop debating it and start building for what comes next.