AI and the future of work: what should you actually do about it?

Practical AI / future of work/7 min read

Strip away the panic and the hype, and every serious analysis of AI and work converges on the same two moves: learn to work with AI, and build a professional reputation under your own name. The first is permanent. The second has a closing window, because within a few years most content will be written with AI and, increasingly, read by AI on your behalf, and both will keep pointing at the people who built their names before the flood.

What does the research actually agree on?

I did an experiment this year. I gave the same question, what should a professional actually do about AI and the future of work, to several AI deep-research tools, each drawing on hundreds of sources, and read the reports side by side. They were produced independently. They came back nearly identical.

Two moves, every time: build AI skills, and build a visible professional brand.

The underlying data explains the consensus. On skills: workers with demonstrable AI skills now command an average wage premium of 62 percent, up from 56 percent a year earlier, and jobs requiring AI skills have grown roughly eight times faster than the overall market since 2019, per PwC's 2026 Jobs Barometer. The World Economic Forum expects 39 percent of core job skills to change by 2030, with roughly 170 million roles created against 92 million displaced.

On visibility, the logic is harsher and less discussed: as AI compresses the value of routine expertise, what remains scarce is trust, and trust attaches to names. PwC's data already shows the shape of it: the fast-growing "professionalised" jobs are the ones where AI does the routine and the human supplies judgement, and judgement is bought from people whose judgement is known.

Two moves. The rest of this article is about why the second one is urgent.

Why do AI skills come first?

Because they are the entry ticket, and the market has stopped being subtle about it. McKinsey's research counted job postings requiring AI fluency growing sevenfold in two years. Microsoft's 2026 Work Trend Index found the number of active AI agents in its ecosystem grew 15 times in a single year, and 65 percent of AI users say they worry about falling behind.

The good news is that the skill is smaller than it looks. It is not machine learning. It is delegation: knowing what to hand to AI, how to brief it, and how to judge what comes back. That is a management skill wearing a technology costume, and it is learnable in evenings, not degrees. If you want the practical version, the whole agents section of this site is that curriculum, starting here: What is an AI agent? A guide for non-technical people.

But AI skills alone have a problem: everyone is acquiring them at once. A premium that everyone earns stops being a premium. Which is what makes the second move the real differentiator.

Why build a personal brand, and why right now?

Because there is a window, and it is closing from two directions at once.

Direction one: everyone is about to write well. When AI makes competent content nearly free, competent stops being worth anything. The feed fills with polished, interchangeable expertise, and the only signal that survives the flood is a recognisable human name with a track record: someone who was visibly building, deciding, and being right or wrong in public. You cannot generate a track record retroactively. It has to be laid down now, in the years when a human voice still stands out.

Direction two: your readers are about to stop reading. This is the part almost nobody prices in.

What happens when AI agents do the reading?

Attention is already being delegated. Gartner projected traditional search volume falling 25 percent by 2026 as queries move to AI assistants, and the shift does not stop at search. The professional who used to scroll the feed increasingly asks an assistant for a digest: what mattered this week, what should I read, who is worth following on this.

Now play that forward. When an agent summarises the week's ideas for its owner, whose ideas make the digest? The agents pull from sources with established authority: the names already cited, already followed, already trusted. And the human on the other end follows the people they were already following before they delegated the reading.

That is the mechanism that closes the window. In the coming attention economy, new voices do not fight for human attention, they fight for machine attention, and machines are trained on who mattered before. The people building recognisable names now are, quite literally, writing themselves into the training data and the recommendation layer of what comes next. The ones who wait will be pitching to an audience that outsourced its curiosity to software with a preference for incumbents.

The window is not mystical. It is the gap between "AI writes most content" (already arriving) and "AI reads most content" (visibly next). Inside that gap, a consistent human voice can still be discovered by humans. After it, discovery itself is machine-mediated.

What does the window plan actually look like?

Unglamorous and doable, which is the pattern with everything real in this field:

Pick one lane and own it in public. Not your whole expertise, one intersection of it, narrow enough that your name and the topic start appearing in the same sentence.

Publish under your own name, on a platform you can be cited from. Digest-writing agents cite sources they can reach and attribute. A named author with a consistent body of work is exactly what they are built to surface.

Show work, not credentials. The flood will be full of polished claims. What survives is evidence of doing: decisions made, things built, numbers shared, mistakes owned. AI can write anyone's opinion; it cannot fake your track record.

Let AI do the leverage, never the voice. Use agents for research, drafts, and distribution, and keep the judgement and the voice recognisably yours, because the voice is the asset the whole plan protects.

Consistency for a few years beats brilliance for a month, and the few years in question are these ones. The professionals who treat 2026 as the visible-building window will spend the 2030s being recommended by other people's agents. The rest will be asking those agents why nobody can see them.

FAQ

What are the two most important career moves in the AI era? Learn to delegate to AI, and build a visible professional reputation under your own name. Independent research analyses converge on exactly these two, and the data backs both: a 62 percent wage premium for AI skills, and a market where trust increasingly attaches to known names rather than credentials.

Is it too late to build a personal brand in 2026? No, but the window is real. Human attention is still partly direct, and a consistent voice can still be discovered. As content creation and content reading both shift to AI, discovery becomes machine-mediated, and machines favour names with an existing track record.

Will AI make personal brands irrelevant? The opposite. When AI makes competent content free, the scarce assets left are trust and judgement, and both attach to people, not posts. AI-mediated reading amplifies this: digest agents surface established, citable names.

How do I start building AI skills without a technical background? Start with delegation, not code: one recurring task handed to an AI agent with a clear brief, reviewed and improved weekly. It is a management skill, and one evening is enough to begin: Build Your First Agent in an Evening.

Author box

Natalya Rojkova is a finance leader who runs capital controls for multi-billion dollar data centre builds by day and builds with AI agents every evening. She writes about delegation, agents, and the new layer of management.

CTA block (Practical AI pillar)

The window is open. I write from inside it. I Hired AI. And you? is my newsletter on AI, work, and building in public: what I test, what breaks, and what the data says. [Subscribe: https://substack.com/@natalyarojkova]

Sources referenced in text (for the page footer, optional)

PwC Global AI Jobs Barometer 2026: 62% wage premium for AI skills (up from 56%); AI-skill jobs growing ~8x faster than the market since 2019; growth of "professionalised" roles.
World Economic Forum, Future of Jobs Report 2025: 39% of core skills to change by 2030; ~170M roles created vs 92M displaced.
McKinsey, November 2025: job postings requiring AI fluency grew ~7x in two years.
Microsoft Work Trend Index 2026: active agents grew 15x year over year; 65% of AI users worry about falling behind.
Gartner: traditional search engine volume projected to fall 25% by 2026 as queries shift to AI assistants.