78 Million New Jobs Are Coming. Here's What They Look Like.

Everyone is wondering where the jobs will go with AI. Start by looking around you.

I was sitting in an AI product demo last month, when my thoughts wandered and I found myself looking around the room, pondering electricity, another general purpose technology.

The projector shone overhead. The HVAC system hummed. The laptop in front of me glowed. The conference room booking system that someone, somewhere, configured. The fiber optic cable feeding the Wi-Fi. The building management software keeping the lights at exactly the right brightness.

So many things in that room existed because electricity was invented as a general purpose technology. Not just the lights. But so many physical items. And I thought about the jobs required to build and support them. The software engineer who built the booking app. The network security analyst protecting the Wi-Fi. The UX designer who made the laptop feel intuitive. None of those jobs existed before electricity reorganized what work meant.

And I thought to myself…in twenty years, someone will sit in a room full of things that don’t exist today, created by people working jobs that don’t have names yet. The same way I couldn’t have predicted cloud architect in 1985, we can’t fully predict what’s coming.

But, based on history, we know it is coming. So, let’s take a moment to suspend the ‘jobs are going away.’ narrative to jump into what the future might hold.

First, an Honest Look at the Data

Before we get to the future, let’s be clear about the present. The World Economic Forum’s Future of Jobs Report 2025 surveyed over 1,000 global employers representing more than 14 million workers. Their headline finding: the WEF estimates a net new change in jobs, not a vanishing of them. By 2030, 170 million new jobs are projected to be created while 92 million are displaced, resulting in a net employment increase of 78 million jobs (roughly 7% of today’s total workforce).

That net positive number is real, and it matters. But two data points in the WEF report deserve a harder look before we move on.

First, the WEF lists Software and Applications Developers and Light Truck Delivery Drivers among the top fastest-growing roles through 2030. I can feel you squinting. If you follow the tech industry, you know junior developers are being laid off right now as AI coding tools eliminate what was once entry-level work. And if you’re watching the transportation sector, autonomous vehicle investment seems to point directly away from more drivers, not toward them.

Here’s how to hold both truths at once. With software, AI automates existing software AND enables the creation of entirely new categories of software products and services that weren’t economically viable before. The market expands. The nature of the work transforms dramatically (less writing boilerplate, more architecting and auditing what AI generates), but net employment grows because there is simply more software in the world to build and maintain. The current junior developer layoff wave is real disruption at the leading edge, not the final destination.

With drivers, the WEF’s 2030 timeframe falls before full autonomous vehicle deployment at scale for complex last-mile delivery. E-commerce growth (itself partly AI-enabled) is currently creating demand that outpaces what technology can absorb in that window. Both things are happening simultaneously, and the 5-year projection reflects that specific window. The 10-year picture may look different.

The broader lesson here is that the labor market transition is nonlinear and full of tensions. Which is exactly why the second WEF data point matters most.

If the world’s workforce was made up of 100 people, 59 would need training by 2030. Of these, employers foresee that 29 could be upskilled in their current roles and 19 could be upskilled and redeployed elsewhere within their organization. However, 11 would be unlikely to receive the reskilling needed, leaving their employment prospects increasingly at risk.

That is the number that should focus every AI transformation leader. Not the net gain of 78 million jobs, but the 11 people in every 100 who fall through the cracks if we don’t act deliberately. (If you want to dig into the specific failure patterns to avoid, I wrote about them in an earlier piece: The AI Digital Rust Belt Is Optional.)

What History Tells Us About Job Evolution

Across every wave of automation in the last century, we’ve seen that when machines take over the doing, humans shift to building and maintaining the machines that do it. When the washing machine took over the laundry, we built and maintained washing machines. When we automated the switchboard, we shifted from manually connecting calls to building and maintaining switchboards.

At FedEx, when their advanced automated sorting system came online, they trained five hundred team members to operate it, including eighty-five new hires whose entire role was to maintain the system. Then they went further. Their BOT-it program turned frontline workers into citizen coders. Employees with no prior coding background developed fifty-six automation products that saved tens of thousands of manual hours. The people who once executed the processes became the people who built the next generation of processes.

This is the through-line. And as AI becomes more capable, the jobs on the building, monitoring, and maintaining side of that equation get more interesting, and more numerous. Let’s get specific about what that actually looks like.

Imagining the Jobs That Come With AI

Here’s the exercise I want you to try. AI is already enabling the design and manufacture of physical products that didn’t exist five years ago. Each of those products requires humans to build, monitor, and maintain the AI systems that create them. Those humans will have job titles we haven’t invented yet.

Consider what’s already emerging from engineering and materials labs:

AI is now designing structural components, including aerospace brackets, architectural beams, car chassis, that look almost biological. Organic curves that eliminate every unnecessary millimeter of material. The shapes are optimized for strength in ways human engineers wouldn’t manually draft.

But someone has to sign off before those parts go into an aircraft. Enter the AI Validation Engineer: an engineer who takes the AI’s design output and stress-tests it against physical reality, because simulation and the real world diverge in ways that experience catches first. They also ensure the part can actually be manufactured by the equipment that exists. We need a host of new jobs to:

• Build the plants to manufacture these new aircraft parts

• Source the materials that go into building the ultra-light parts

• Maintain the robots that assemble next-generation parts

AI is discovering new battery chemistries by simulating millions of atomic combinations. The Crystal Structure Validation Engineer is the person who takes those AI-discovered compounds and tests them across real-world temperature ranges, charge cycles, and edge conditions that the model never saw. The AI finds the candidate. This person determines whether physics cooperates.

AI is enabling responsive meta-materials, including fabrics and building materials engineered at the microscopic level to respond to environmental stimuli such as changing their thermal insulation, absorbing specific sound frequencies, adapting to conditions. Someone has to define those stimulus-response parameters. Along with this new technology comes new roles in research, manufacturing, and maintenance.

As we explore drone delivery services we see the emergence of Drone Mission Controllers responsible for coordinating the intersection of drones between services and other flying objects operating in an unscripted environment.

AI is enabling screenless environmental sensors: earpieces, glasses, pendants with outward-facing AI chips that continuously parse your physical surroundings. Someone has to define what that AI pays attention to and what it ignores. The Contextual Awareness Designer is responsible for that model (as well as for the moment when the system misclassifies something it shouldn’t). They’re part product designer, part cognitive scientist, part safety engineer.

Not one of these job titles appears in a standard HR system today. Every single one maps directly to the build-monitor-maintain pattern. And every single one requires a human who deeply understands both the technology and the domain it’s operating in.

The Organizations That Can See What’s Coming

The above roles don’t emerge by accident. They are invented by organizations that have built the capability to sense what’s needed next, form teams to define it, and develop people to fill it faster than a competitor can recognize the need exists.

The Hyperadaptive Model sets the stage for this exact outcome. By integrating learning loops and changing the operating model, organizations move their people off career ladders and into a network that evolves roles continuously and intentionally, rather than scrambling when the disruption has already arrived.

As Hyperadaptive notes, the goal in the most advanced stages of organizational AI maturity isn’t to fill today’s roles better. It’s to build an organization that can sense the need for a new role, define it, and empower people to fill it faster than anyone else.

The new jobs are coming. The question is whether your organization will be inventing them, or scrambling to staff them after you’ve laid off all of your best workers or before someone else does.

One Last Thought

I left that AI session thinking about all of it. The room full of things that wouldn’t exist without electricity. The jobs that electricity created. The things that are being built right now in labs and foundries that will require people to build them, watch over them, calibrate them, and course-correct them when they stray.

The piece of this I keep coming back to is imagination. Because the organizations that will define this decade aren’t the ones slashing headcount. They’re the ones actively inventing what comes next.

That’s a different kind of leadership. And it’s available to anyone willing to start imagining.

---

If you’re building the organizational capability to navigate this transformation, Hyperadaptive by Melissa Reeve (IT Revolution Press, 2026) offers a research-grounded roadmap — from the first AI experiments to the fully AI-native enterprise. Pre-order now.

Related Article

---

SPECIAL OPPORTUNITY

The first paid community peer roundtable is coming up on April 16th: Getting Your Organization to Agree on What 'Adopting AI' Means. RSVPs are already rolling in, and I'm genuinely excited about the group that's forming. If you've been thinking about joining the paid side of Hyperadaptive, this is a great reason to do it. Small room, real talk, no fluff. Upgrade to paid to access.