How to AI-Proof Your Career

Ask whether AI will take a given job and you'll usually get a confident answer in one direction or the other. Michael Domanic, Head of AI at UserTesting, gives the more honest one. Asked directly, he says "we don't know," framing that as both the bad news and the good news, an argument for facing the uncertainty honestly rather than defaulting to reassurance or alarm. He also pushes back on the comparison most AI leaders reach for, the Industrial Revolution, pointing out that period was roughly 80 genuinely painful years for the people living through it, not a clean, comfortable transition.

That honesty is actually the right starting point for thinking about how to AI-proof your career. Nobody can promise a specific role is safe. What's possible is building the specific traits that keep showing up, across very different companies and industries, as the ones AI isn't replacing.

Learn what companies are actually hiring for now

Brad Axen, who built the open-source agent framework Goose at Block, describes a concrete shift in how his own team hires, and it's a clear signal for anyone thinking about how to AI-proof your career. He no longer runs "how well do you know Python" interviews, something he did as recently as about ten years ago when he joined. Interviews now center on problem-solving with full AI tool access. But the shift isn't toward hiring generalists who are simply comfortable with AI. Axen is explicit that the company still wants senior domain experts capable of catching an AI's incorrect outputs, not just AI-fluent generalists.

That distinction matters. The safest position right now isn't being the person who can prompt an AI tool well. It's being the person with enough real domain expertise to know when the AI's answer is wrong. AI fluency is becoming table stakes. Deep, verifiable expertise in a specific domain is what's actually scarce.

Find your zone of genius before AI finds it for you

Jason Li, CTO of the legal and accounting timekeeping company Laurel, offers a genuinely practical way to apply this to your own role. Laurel's product generates granular data on what knowledge workers actually spend their time on, and Li argues most companies default to standardized, cookie-cutter role definitions purely because they're easier to manage, not because they reflect where someone actually adds the most value. With real data, a company might discover that one person is an exceptional specialist in a narrow part of their role, and should be freed to focus almost entirely on that rather than splitting time across a generic job description.

You don't need Laurel's software to apply the same logic to yourself. Honestly audit where your work actually produces outsized results versus where you're just filling time on tasks anyone, or any tool, could do adequately. The parts of your role where your specific judgment consistently outperforms a generic approach are the parts worth deliberately deepening. The rest is exactly the kind of work AI is already absorbing.

Climb toward the purpose of your role, not just the tasks inside it

An Adobe partnerships executive relayed a story from NVIDIA CEO Jensen Huang worth sitting with directly. Huang used radiology as a case study: despite predictions that AI-assisted scan reading would eliminate radiologist jobs, there are actually more radiologists today, because reading a scan was never the actual purpose of the role. The purpose was disease detection. AI absorbed the mechanical task and elevated people toward the higher-purpose work the task was always in service of.

This is one of the clearest tests for how to AI-proof your career at the task level. Ask what the actual purpose behind your day-to-day tasks is, not just the tasks themselves. If AI is automating the mechanical layer sitting on top of a real purpose, deepen your ability to do the purpose-level work underneath it. If there's no purpose layer left once the task is automated, that's a genuine signal to move toward a role where there is one.

Build the layer that stays stubbornly human

Aaron, who works in enterprise AI marketing at Adobe, makes a prediction worth building directly into any AI-proofing strategy: the future of work will involve less screen and machine interaction, not more. Her reasoning is that if AI can genuinely understand a person's context and objectives, it frees up time specifically for relationship-building, trust, and intimacy, traits she frames as permanently non-automatable, and as the real differentiator once AI makes comparable products and speed available to everyone.

She offers a concrete formula for this worth internalizing directly: trust equals reliability times intimacy times credibility, divided by self-orientation. It's a useful personal audit. The parts of your work that build genuine trust through consistency, closeness, and credibility, rather than raw output, are the parts least exposed to automation, almost by definition, since those qualities require an ongoing relationship, not a single deliverable.

Use a simple personal test to check you're actually getting stronger

Rana Gujral, founder of the voice-AI company Behavioral Signals, offers a genuinely reusable framework for anyone wondering whether their own AI use is building their career or quietly eroding it. His test is two direct questions: does this make me more capable when I'm using it and also when I'm not, and am I becoming more myself through this, or less. He explicitly distinguishes reversible augmentation, where stepping away from the tool still leaves you capable, from replacement, where stepping away leaves a hole because the tool became load-bearing for your competence or identity.

Run your own AI usage through this test honestly. If a skill only exists while the tool is open, and disappears the moment you close it, that's not AI-proofing your career. It's quietly outsourcing the exact capability that was supposed to make you more valuable.

Don't mistake tool fluency for the whole skill set

Tudor Achim, CEO of the formal-math AI company Harmonic, offers a useful, somewhat contrarian check on this entire conversation. He argues explicitly against treating AI literacy as a new core curriculum priority, even in education, contending that the bigger blockers to good outcomes are usually mundane and non-AI, and that genuine critical-thinking skills need to already be in place before AI becomes useful on top of them, not instead of them. Scott Likens, PwC's Chief AI Engineer, points to a concrete real-world version of this same idea: a school called Alpha School in Austin that rebuilt its curriculum around AI-assisted academic learning in the morning paired with afternoon project-based human skills work, fixing a bike, gardening, working across age groups.

The lesson for how to AI-proof your career is the same at the adult, professional level. Learning to prompt well is a surface skill. The foundational judgment, critical thinking, and hands-on problem-solving underneath it are what actually make that surface skill useful, and they're not something AI tool fluency alone can substitute for.

Expect the hardest shift to be psychological, not technical

Roy Mann, CEO of monday.com, shared an internal story that illustrates the last piece of this clearly. A developer on his own team was initially terrified watching an AI agent implement and deploy UI fixes faster than he could read the change requests. Within a week, the same employee shifted from feeling he was competing with the agent to feeling he now operated it. Mann attributes that shift specifically to a change in how the employee positioned himself relative to the process, not to any change in the technology over that week.

That shift, from feeling replaced by a tool to feeling like its operator, is arguably the real work behind how to AI-proof your career. The technical skill of using AI well tends to come quickly. Repositioning yourself as the person directing and verifying the work, rather than the person doing the task the tool now does, is the harder and more durable change.

Keep learning from how real careers are actually adapting

Nobody has a complete, verified answer for how to AI-proof your career, including the people building these systems. What they have are real, tested patterns worth learning from directly. Subscribe to The AI Report for ongoing, firsthand coverage of how people are actually navigating this, straight from the people living it.

‍