Embracing the Zigzag: Why Generalists May Thrive in the AI-Driven Future of Work
Last year, while catching up with an old friend over a round of golf, our conversation took an interesting turn. Swapping life updates, he laughed and said, “I genuinely don’t know what to call you anymore. Are you still in finance? Teaching? Running the school? Writing? Pick a lane, mate.” I laughed too, but his comment lingered with me.
For much of my adult life, I’ve carried what I once thought was a slightly embarrassing label: jack of all trades, master of none. I’ve moved from finance to teaching, then to managing a language school, running a couple of small businesses, and now writing. My career path has been anything but linear—full of twists and turns.
However, the world is evolving in ways that make this zigzagging less a sign of indecision and more of an asset—especially in certain fields. What follows is my reflection on this shift, with an awareness that my perspective is shaped by my own diverse experiences. So take it with a grain of salt.
The Era of the Deep Specialist Looks Different Than It Used to
I grew up believing the career path was straightforward: pick a field, become an expert, stay in your lane, and climb the ladder. This was the blueprint for my parents’ generation, and it largely worked for them. But the foundations of that deal have shifted.
The World Economic Forum’s Future of Jobs Report 2025 quantifies this change. It predicts that by 2030, approximately 39% of workers’ core skills will be transformed or become outdated.
This isn’t about robots taking over jobs per se; it reflects the shortening half-life of expertise across many fields. While some professions—medicine, engineering, classical music—still reward decades of deep practice, an increasing portion of white-collar work demands continual refreshment of knowledge.
If your professional identity hinges on a single tool or job title within a rapidly evolving field, you’re in a more vulnerable position than two decades ago.
AI Competes Hardest with Narrow, Well-Defined Work
When I was younger, I assumed specialization equaled job security. Become an expert, and you become irreplaceable. But AI has complicated that assumption, especially for routine tasks.
Generative AI excels at work confined to a single, well-defined domain—drafting standard contracts, writing routine code, producing first-draft marketing copy, or extracting structured analysis from spreadsheets. The more pattern-based and narrow the task, the more AI can encroach on it.
Anthropic’s Economic Index, which tracks real-world use of the AI assistant Claude, confirms this trend: most applications currently cluster around software development and technical writing, areas rich in narrow, technical, and repetitive tasks.
Similarly, the WEF report highlights that two-thirds of employers plan to hire specifically for AI skills, while 40% intend to reduce headcount where AI can automate tasks.
That said, AI also augments productivity for many specialists—the radiologist analyzing scans faster, the lawyer completing document review in an afternoon, or the developer shipping features more quickly. Deep expertise combined with AI tools remains a formidable advantage. The risk lies in narrow experts unwilling to adapt their skills or leverage AI.
What AI struggles with is the messy in-between: connecting disparate fields, spotting how patterns from biology apply to logistics, or recognizing when data misleads because of personal experience with the people behind it. This kind of cross-domain insight is where generalists often excel.
The Case for the Curious Career-Jumper
David Epstein’s Range: Why Generalists Triumph in a Specialized World argues that in complex, fast-changing environments, breadth often outshines depth. As he writes, “The more constrained and repetitive a challenge, the more likely it will be automated.”
This aligns with the WEF’s identified core skills for 2025: analytical thinking, resilience, flexibility, agility, leadership, creative thinking, motivation, and self-awareness. None are tied to a single domain, but all favor those who have repeatedly started from scratch.
Why the Self-Teachers May Have an Unfair Advantage
Formal education remains valuable, and I’m not dismissing degrees. But curricula often reflect what mattered several years before graduation. In contrast, the self-taught learn driven by curiosity, not obligation.
They dive into YouTube tutorials, explore rabbit holes on Reddit, build small projects, and engage with AI tools to deepen understanding. This informal, continuous learning embodies the curiosity and lifelong learning identified by the WEF as critical skills for the future.
I see this pattern in myself. I never formally studied writing; most of my skills come from reading admired authors, imperfectly imitating them, receiving feedback, and iterating. The same applies to my other careers—finance was a trial by fire, running a school required figuring things out on the fly. Being the new person, often the least experienced, builds a unique confidence:
“I can probably learn this,” rather than, “I already know this.”
In an economy where how fast you can relearn matters more than what you already know, that mindset is invaluable.
The Bottom Line
I’m not suggesting you abandon your specialty tomorrow. The most effective professionals I know combine deep expertise with a willingness to explore outside their core domain. Specialists will continue to thrive—especially those who embrace AI as a tool, not a threat.
But here’s the tougher question I keep returning to: the generalist path carries a structural cost. It means repeatedly being the least competent in the room, never quite finishing the climb peers are making, and resetting often. Retroactive vindication is easy; living with the uncertainty is harder.
The real skill is comfort with uncertainty—the willingness to zigzag into an unmapped future without guaranteed payoffs. Everything else is just storytelling.
If you’re looking for permission to zigzag, this isn’t it. But if you want a prompt to reflect on your own path, maybe it is.
Source: Here
