What makes us human when machines can do most of our work?

Your teenager asked what your job will be in ten years, and you realised you gave them a confident answer you do not actually believe. The words came out fine. The voice was steady. But somewhere behind your own sentence you heard a gap open, the distance between what you said and what you actually hold to be true.

Here is the answer worth holding, and it is steadier than the one you improvised over dinner. Being human in the age of AI comes down to three things a machine cannot stand in for: judgement when the situation is truly ambiguous, creativity that originates rather than recombines, and presence, the full attention of one person on another. Machines do most of the work. You do the part that decides whether the work was worth doing. That is not a comfort. It is a job description, and it is yours to grow into and own.

Why does the confident answer feel hollow now?

Because you are not the only one rewriting it. In a December 2025 Penn Foster survey of 506 parents of 14-to-18-year-olds, 55% said they feel anxious about whether their child is being prepared for the jobs that will exist in an AI-driven future, while another 33% said they are unsure how to feel. In the same survey, 37% said AI has changed the expectation that their child would pursue a four-year degree. Read that figure again. Parents are actively rethinking the path they once took for granted while privately confessing they do not know what to think. That is the exact shape of the moment at your own dinner table: the answer given, and the belief actually held, no longer the same answer.

The teenagers feel it too. In a 2025 Junior Achievement survey of 1,000 American teens aged 13 to 18, 57% said the rise of AI had negatively shaped how they view their future careers, even as 94% stayed optimistic about their prospects. They are not naive. They are carrying a hope and a fear at the same time, which is precisely what you were doing when you smiled and gave the confident reply.

What is the work that stays mine?

The instinct is to answer with a list of tasks. That is the wrong altitude. The work that stays yours is defined by the human operating system you run (the way you think, regulate yourself, and decide), not by the tasks AI has not reached yet. This sits inside a larger arc. Phase One, the Age of Effort: work hard, get a little more, linear growth. Phase Two, the Age of Scale: build once, sell to millions, exponential growth. Phase Three, the Age of Acceleration: output decoupled from human effort almost entirely, the phase AI unlocks. We are crossing into Phase Three now, and it is precisely there that the human layer becomes the thing that decides the outcome.

Look at what the tools actually do to capable people. In a field experiment with 250 employees at a technology consulting firm, published in the Journal of Applied Psychology in 2025, access to generative AI raised creativity, as scored by supervisors and external evaluators, only for those with strong metacognitive skills (the capacity to think about your own thinking: to plan an approach, monitor it as you go, and revise the strategy). For employees low in metacognition, the creativity gain was negligible. The same tool, the same prompt box, two completely different outcomes, decided entirely by the human thinking in front of it. AI is not a plug-and-play creativity solution. It is an amplifier, and an amplifier returns what you feed it.

Boston Consulting Group put a number on the same truth from the boardroom. Drawing on a survey of 1,000 senior executives across 59 countries, BCG advises that successful AI work splits roughly 70% to people and process, 20% to data and technology, and only 10% to the algorithms themselves. In that same research, 74% of companies were struggling to achieve and scale value at all, and just 4% had built advanced capabilities that consistently generate substantial value. The bottleneck is no longer the technology. It sits exactly where it has always sat: in the human capacity to decide, to align, and to lead.

So which capabilities actually rise as the machines take more?

The OECD answered this with hard labour-market data, not opinion. In its 2025 report Bridging the AI Skills Gap (the OECD is the Paris-based economic body that tracks work and skills across its member countries), it found that roughly one in three job vacancies now sits in occupations highly exposed to AI, and that the cognitive skill whose demand has risen most across those jobs is originality. The World Economic Forum's Future of Jobs Report 2025, drawing on over 1,000 employers who together represent more than 14 million workers, points the same way: analytical thinking tops the list of core skills, called essential by roughly seven in ten companies, followed by resilience, flexibility and agility, then leadership and social influence, with creative thinking ranked next and among the fastest-rising. Notice what every one of those is. None is a tool. Each is a property of the person. They are the things you bring to the machine, the judgement that frames the question and the discernment that weighs the answer.

Why does presence belong on a list about work?

Because your state decides your judgement, and judgement is the thing you are now paid to bring. Consider one of the cleaner pieces of recent evidence. Researchers analysed nationally representative records of Australian general practitioners' consultations in a 2024 study in Medical Decision Making, and found that as the working day wore on, doctors grew steadily more likely to make the easier, less effortful prescribing choice, the pattern that marks decision fatigue. Same doctors, same guidelines, same kinds of patients. What changed was the human being's depleted capacity to do the harder thing, so they reached for the easier one.

That is not a story about one profession. It is a story about the operating system underneath every decision. The autonomic research bears it out. A 2024 review in the journal Neuroscience, mapping what is now called the brain-heart connection, confirms that higher resting heart rate variability (the natural beat-to-beat variation in your pulse, a marker of how well your nervous system is regulated) tracks with stronger executive function and more activity in the prefrontal regions that handle attention and self-control. Your physiology shapes your capacity to think. Which means presence, regulation, and coherence (head-and-heart alignment) are not soft additions to a leader's day. They are the conditions under which your one irreplaceable contribution, judgement, can actually surface. Most leaders are trying to install new software on broken hardware. The upgrade that matters now is the one you make to yourself.

So here is the answer you can give your teenager and mean it. The work that survives is the work that requires a person to be fully present, to think about their own thinking, to originate rather than recombine, and to choose well when no rule fits the situation. Phase One was muscle. Phase Two was machine. Phase Three is mind. The leaders who win the next decade are the ones who upgrade themselves first, and so are the children we are raising to follow them.

How do I start building this, today?

1. Name the gap out loud. Write down the confident answer you gave and the belief you actually hold. The distance between them is your work, made visible.
2. Strengthen metacognition before reaching for tools. Before any AI task, decide your approach, watch how it goes, and revise it. The tool rewards the person who thinks about their thinking.
3. Protect the conditions for judgement. Schedule your hardest decisions for when you are resourced, not depleted, and treat presence as a performance variable.
4. Redirect the dinner-table answer toward capability. Tell your teenager what humans are becoming better at, not only what machines are taking. Point the energy at what they get to build.

About the author

Thomas Green

British technology futurist, AI keynote speaker and advisor. Thirty years across enterprise technology and AI strategy, helping leaders navigate the future of work. The futurist who died.