--- title: "What One CEO Learned After a Year of Replacing His Own…" canonical: "https://www.metaintro.com/blog/ceo-year-replacing-own-work-with-ai-lessons-2026" language: "en" author: "drashtigarach" published: "2026-06-09T16:58:22.000Z" modified: "2026-06-09T16:58:22.000Z" --- [Back to Blog](/blog) [AI](/blog/tag/ai)[Career](/blog/tag/career)[Technology](/blog/tag/technology)[Self Help](/blog/tag/self-help)[Career Tips](/blog/tag/career-tips) # What One CEO Learned After a Year of Replacing His Own Work With AI One CEO spent a year handing his own job to AI. See the 3 tasks it took, the work it never could, and how to future-proof your career now. [![Drashti Garach](https://cdn.metaintro.com/rs:fill:40:40/q:72/plain/images/5719d740-e510-42bc-8017-e040d145f35f_1766029465094.png)Drashti Garach @DrashtiGarach](/blog/author/drashtigarach) [June 9, 2026](/blog/archive/2026/06)10 min read ![What One CEO Learned After a Year of Replacing His Own Work With AI](https://cdn.metaintro.com/rs:fill:1200:675/q:78/plain/images/ceo-year-replacing-own-work-with-ai-2026-hero.png) [https://x.com/intent/tweet?text=What%20One%20CEO%20Learned%20After%20a%20Year%20of%20Replacing%20His%20Own%20Work%20With%20AI&url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026](https://x.com/intent/tweet?text=What%20One%20CEO%20Learned%20After%20a%20Year%20of%20Replacing%20His%20Own%20Work%20With%20AI&url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026)[http://www.facebook.com/sharer.php?u=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026](http://www.facebook.com/sharer.php?u=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026)[https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026&title=What%20One%20CEO%20Learned%20After%20a%20Year%20of%20Replacing%20His%20Own%20Work%20With%20AI](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026&title=What%20One%20CEO%20Learned%20After%20a%20Year%20of%20Replacing%20His%20Own%20Work%20With%20AI)[mailto:?subject=What%20One%20CEO%20Learned%20After%20a%20Year%20of%20Replacing%20His%20Own%20Work%20With%20AI&body=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026](mailto:?subject=What%20One%20CEO%20Learned%20After%20a%20Year%20of%20Replacing%20His%20Own%20Work%20With%20AI&body=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026) When a CEO spends a full year deliberately handing his own job to AI, the takeaway is not that machines run companies now. It is sharper and more useful than that. The drafting, the summarizing, and the number-crunching moved to AI fast, while the judgment calls, the hard conversations, and the trust between people stayed stubbornly human. At [Metaintro](https://www.metaintro.com), we read these experiments so job seekers can act on them, and the lesson here is simple. The safest roles in 2026 are the ones built on the work AI still cannot do. This piece breaks down which tasks fell to AI, which held, and how to position your own role before the choice gets made for you. ## What did the CEO actually hand over to AI? The experiment, documented by [Axios](https://www.axios.com/), followed a leader who treated himself as the test subject. Instead of asking his staff to adopt AI first, he ran his own calendar, inbox, and output through the tools for roughly a year to see what would break. The pattern that emerged matches what larger studies keep finding. The first work to go was repeatable and verifiable. Drafting memos, summarizing long reports, turning rough notes into clean slides, and pulling numbers into a first-pass analysis all moved to AI quickly. These are tasks with a clear input and a checkable output, which is exactly the kind of work the technology handles well. This tracks with the [World Economic Forum](https://www.weforum.org/publications/the-future-of-jobs-report-2025/) Future of Jobs Report 2025, which found that employers estimate 47% of work tasks are still performed primarily by humans today, 22% are handled mainly by technology, and 30% involve a mix of both. By 2030, the Forum expects those shares to move toward a near-even split. The CEO's year was a fast-forward through that shift inside a single person's job. Crucially, the experiment did not end with him out of a role. He still made most of the same calls he would have made anyway. What changed was how much grunt work surrounded those calls, and how visible the boundary became between tasks and trust. ## Which tasks did AI take first, and why? If you want to predict what AI takes from your own job, look for three traits. The task is repeatable, the output is easy to verify, and the work does not depend on a relationship. The CEO's drafting and analysis checked all three boxes, so they went first. The numbers from enterprise research point the same direction. The [IBM](https://newsroom.ibm.com/2026-05-04-ibm-study-ceos-are-reshaping-c-suite-roles-for-the-ai-era) 2026 CEO Study, based on 2,000 global CEOs, found that leaders expect 48% of operational decisions to be made by AI where guardrails apply by 2030. Note the qualifier. AI gets the decisions that fit inside clear rules and known patterns, not the ones that require reading a room. That same [IBM study](https://www.ibm.com/thought-leadership/institute-business-value/en-us/c-suite-study/ceo) found that 29% of employees are expected to need reskilling into different roles between 2026 and 2028, while 53% will need upskilling to do their current jobs better. In other words, most workers are not being replaced outright. They are being asked to drop the tasks AI absorbs and pick up the work it cannot. The [Future of Jobs](https://www.weforum.org/stories/2025/01/future-of-jobs-report-2025-jobs-of-the-future-and-the-skills-you-need-to-get-them/) data adds another useful number. Employers expect 39% of the core skills workers need will change by 2030, with analytical thinking, resilience, flexibility, and AI literacy rising fastest. The skills that hold value are the ones the CEO found AI could not cover. ## What work could the AI not do? This is where the year-long experiment earned its keep. The CEO kept hitting a wall on the same category of work, and it was not technical. It was human. AI could draft a difficult message, but it could not sense the emotional weight of delivering one. It could lay out the options for a promotion, but it had no read on the office politics around the decision. As the [CEOWORLD magazine](https://ceoworld.biz/2026/05/28/why-ai-cannot-replace-leadership-judgment/) analysis of leadership judgment put it, AI gives you an answer with no sense of the human context. It cannot feel the fragility of a struggling employee's confidence or the stakes hiding behind a routine-looking call. There is a deeper risk too. AI tends to make suggestions confidently even when they are shallow or wrong. A [Medium analysis](https://lucavettor.medium.com/ai-can-execute-your-work-but-it-cannot-own-your-judgment-23e16ff3c9fd) by one strategist framed it cleanly. AI can execute your work, but it cannot own your judgment. The danger is not just that AI gets things wrong. It is that it can lull you into skipping the second look you would normally take, which dulls your own judgment over time. Then there is the relationship cost. [Fortune](https://fortune.com/2026/03/28/social-offloading-ai-boss-empathy/) reported on a phenomenon researchers call social offloading, where people outsource the human parts of work to AI. If you always ask a tool how to respond to your boss, you never actually learn to engage with your boss or build that relationship yourself. The CEO's experiment surfaced the same trap. The more he let AI handle the interpersonal scaffolding, the weaker those muscles got. A useful way to make sense of all this is the split between tasks and trust. As one [MIT Sloan Management Review](https://sloanreview.mit.edu/article/when-not-to-use-ai/) piece on when not to use AI argues, tasks are repeatable processes that benefit from speed, while trust is the human currency of management. Use AI on tasks. Protect the trust. ## What does this mean for your career? Here is the part you can act on this week. The CEO's year of self-experimentation is really a map of where to stand so the next round of AI adoption works for you instead of around you. First, own the final call. AI can hand you ten options and a confident recommendation, but someone has to decide which one fits the real situation, and that someone should be you. Build a reputation as the person whose judgment is trusted, not the person who produces the most first drafts. The [WEForum](https://www.weforum.org/stories/2025/12/work-transformation-skills-agility-growth/) work-transformation research is blunt that judgment and decision quality rise in value as automation spreads. Second, own the relationships. The tasks that survived the experiment were the ones rooted in trust, such as delivering hard feedback, managing a tense negotiation, mentoring someone through a setback, and reading the politics of a room. These are not soft extras. They are now the durable core of a job. If your current role is mostly solo execution, find ways to attach yourself to client relationships, cross-team coordination, or people leadership. Third, become the person who directs AI rather than competes with it. The [IBM](https://hrexecutive.com/half-of-ceos-want-new-ai-roles-what-ibm-says-hr-can-do-to-prepare/) research shows roughly half of CEOs are creating entirely new AI-related roles, and the workers who can frame a problem, prompt well, and then check the output against reality are the ones filling them. Treat AI fluency as a baseline skill the way spreadsheets once were. Fourth, audit your own job honestly. List your weekly tasks and mark each one as repeatable-and-verifiable or judgment-and-trust. The first column is your exposure. The second is your moat. Spend the next quarter deliberately growing the second column, whether that means volunteering for a stakeholder-facing project or taking on a mentee. For a structured starting point, the way you present that shift matters, and a sharper resume that foregrounds judgment and outcomes over task lists is part of getting hired in this market through [Metaintro](https://www.metaintro.com). The workers who thrive after this kind of experiment are not the ones who avoid AI. They are the ones who let it take the tasks and then double down on the trust. ## How fast is this shift actually moving? Fast enough to matter for your next job move, but not so fast that the human work disappears. The [WEF press](https://www.weforum.org/press/2025/01/future-of-jobs-report-2025-78-million-new-job-opportunities-by-2030-but-urgent-upskilling-needed-to-prepare-workforces/) release on the Future of Jobs Report projects 170 million new roles created and 92 million displaced by 2030, a net gain of 78 million jobs. The churn is real, and the safe ground is the work the CEO could not automate. It also matters how employers are responding. In the [Future of Jobs](https://www.weforum.org/stories/2025/01/future-of-jobs-report-2025-jobs-of-the-future-and-the-skills-you-need-to-get-them/) data, most employers say they plan to reskill and redeploy their people to work alongside AI rather than replace them outright, while a far smaller share expect to cut roles where automation takes over. The story is less about elimination and more about reshaping. The tasks shrink, the judgment grows, and the worker who notices first wins. That is the whole point of running a year-long experiment in public. You get to learn the lesson without losing the job. Lacey Kaelani, CEO of [Metaintro](https://www.metaintro.com), told [Nation's Restaurant News](https://www.nrn.com/quick-service/from-burger-king-to-taco-bell-how-ai-employee-assistants-could-quietly-reshape-restaurant-labor): "There's a major difference between AI that assists in scheduling hourly workers, a significant issue, and AI that rates an employee's politeness." That line captures the boundary the CEO kept running into all year. AI is genuinely useful at the mechanical work of scheduling, sorting, and drafting. It becomes risky the moment it tries to judge the human side of a job, the politeness, the effort, the intent. The workers who understand that line, and plant themselves on the human side of it, are the ones building careers that last. --- ## Related Articles - [New Research Reveals the Surprising Ways People Actually Use AI at Work](https://www.metaintro.com/blog/how-people-really-use-ai-at-work-2026) - [The Real Reason Workers Push Back on AI Isn't the Technology](https://www.metaintro.com/blog/real-reason-workers-resist-ai-not-technology) - [Software Engineers Are Becoming AI Agent Managers, How to Stay Ahead of the Shift](https://www.metaintro.com/blog/software-engineers-ai-agent-managers-how-to-stay-ahead) - [Why AI Could Create More Jobs Than It Eliminates by 2028](https://www.metaintro.com/blog/why-ai-could-create-more-jobs-than-it-eliminates-by-2028) - [Why AI Does Not Have to Mean Layoffs for Your Job in 2026](https://www.metaintro.com/blog/why-ai-does-not-have-to-mean-layoffs-for-your-job-2026) - [How a Confused AI Rollout Is Hurting Companies and Staff](https://www.metaintro.com/blog/confused-ai-rollout-hurting-companies-staff) - [Verizon's CEO Says AI Will Replace Most Customer Service Jobs, Here's Who's Exposed](https://www.metaintro.com/blog/verizon-ceo-ai-replace-customer-service-jobs-whos-exposed) - [Fed Warns AI's Job Costs May Hit Faster Than Its Benefits](https://www.metaintro.com/blog/fed-warns-ai-job-costs-may-hit-faster-than-benefits) - [How to Tell If AI or Cost-Cutting Is Really Behind Tech Layoffs in 2026](https://www.metaintro.com/blog/how-to-tell-if-ai-or-cost-cutting-behind-tech-layoffs-2026) - [How Box Created 13 Brand-New Jobs Because of AI](https://www.metaintro.com/blog/how-box-created-13-brand-new-jobs-because-of-ai) --- ## People Also Asked ### Q: Will AI replace my job in 2026? A: For most workers, no. The bigger pattern is task replacement rather than job replacement. AI absorbs the repeatable, verifiable parts of a role first, such as drafting and basic analysis, while the [IBM](https://newsroom.ibm.com/2026-05-04-ibm-study-ceos-are-reshaping-c-suite-roles-for-the-ai-era) study shows most employees are being reskilled or upskilled rather than removed. The risk concentrates in jobs that are almost entirely repeatable execution with no judgment or relationship component. ### Q: Which work skills are safest from AI? A: Judgment, relationship building, and accountability rank highest. The [WEForum](https://www.weforum.org/stories/2025/12/work-transformation-skills-agility-growth/) research and the CEO's own experiment both point to the same conclusion, that work requiring human context, trust, and a final decision holds its value while routine output does not. Analytical thinking and AI literacy round out the durable list. ### Q: How do I make myself valuable in an AI workplace? A: Position yourself as the person who directs AI and owns the call, not the one who competes with it on output speed. Audit your tasks, shed the ones AI can verify, and grow the ones rooted in trust and decision-making. Pairing that with a resume that leads with outcomes and judgment, supported by tools like [Metaintro](https://www.metaintro.com), helps you show employers the work AI cannot do. Future-proof your career by getting ahead of the shift instead of reacting to it. The CEO who spent a year handing his work to AI came out knowing exactly where humans still win, and you can apply that map to your own role today. Build the judgment, own the relationships, and learn to direct the tools, then put it all in front of employers who are hiring for exactly that. Start with [Metaintro](https://www.metaintro.com/signup). ### Share this article [https://x.com/intent/tweet?text=What%20One%20CEO%20Learned%20After%20a%20Year%20of%20Replacing%20His%20Own%20Work%20With%20AI&url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026](https://x.com/intent/tweet?text=What%20One%20CEO%20Learned%20After%20a%20Year%20of%20Replacing%20His%20Own%20Work%20With%20AI&url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026)[http://www.facebook.com/sharer.php?u=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026](http://www.facebook.com/sharer.php?u=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026)[https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026&title=What%20One%20CEO%20Learned%20After%20a%20Year%20of%20Replacing%20His%20Own%20Work%20With%20AI](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026&title=What%20One%20CEO%20Learned%20After%20a%20Year%20of%20Replacing%20His%20Own%20Work%20With%20AI)[mailto:?subject=What%20One%20CEO%20Learned%20After%20a%20Year%20of%20Replacing%20His%20Own%20Work%20With%20AI&body=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026](mailto:?subject=What%20One%20CEO%20Learned%20After%20a%20Year%20of%20Replacing%20His%20Own%20Work%20With%20AI&body=https%3A%2F%2Fwww.metaintro.com%2Fblog%2Fceo-year-replacing-own-work-with-ai-lessons-2026) For job seekers ## The job you want exists. We’ll find it. No more scrolling job boards. Get matched roles by text, with the pay and the reason they fit. [Get Started Free](/signup)[Search matching jobs](/jobs/search) - MatchCompare live roles against your current evidence. - PositionTurn proof projects into role-specific applications. - ImproveUse market feedback to keep the skill plan current. [Return to navigation](#main-navigation)