How AI Is Actually Changing Work (Not How You Think)

The narrative is wrong: AI is not replacing jobs — it is reshaping them

The most common question professionals ask about AI is: will it take my job? The evidence from 2024-2026 tells a more nuanced story. AI is not eliminating jobs wholesale — it is transforming what those jobs look like, often in ways that create more value and more interesting work.

McKinsey's 2024 Global Survey on AI found that 72% of organisations have adopted AI in at least one business function, up from 55% in 2023. But the same research shows that only 8% of those organisations reduced headcount as a result. The far more common outcome: existing employees were retrained to work alongside AI, and new roles were created that did not exist before.

The pattern is consistent across industries. AI automates the repetitive, predictable parts of a role — data entry, first-level customer queries, routine reporting, document summarisation. What remains is the work that requires human judgment: building relationships, navigating ambiguity, communicating across teams, making decisions with incomplete information. These are the skills that define career advancement, and they are becoming more valuable, not less.

Jensen Huang, CEO of NVIDIA, put it bluntly in a March 2025 CNBC interview: "AI layoffs signal weak leadership. Firms with imagination do more with more, not less." He added: "AI elevates workers. Every carpenter or plumber could now be an architect." NVIDIA itself is still hiring aggressively — a signal that the company building the infrastructure of AI does not believe AI replaces people.

The IKEA story: How AI created a billion-dollar division instead of eliminating jobs

The most powerful case study for AI-augmented work comes from an unlikely source: IKEA. In 2024, IKEA introduced an AI chatbot to handle customer service inquiries. The results were immediate — the bot resolved 60% of all customer tickets without human intervention. Faster response times, lower cost per interaction, higher customer satisfaction for routine questions.

The typical corporate playbook at this point would be to reduce customer service headcount. IKEA did the opposite. Their leadership looked at the remaining 40% of unresolved tickets — the ones the AI could not handle — and asked a different question: what are these customers actually looking for?

The answer was surprising. A significant portion of those unresolved queries were not complaints or product issues. They were customers asking for interior design advice. People wanted help putting rooms together, choosing colour palettes, combining products into cohesive living spaces. This was a need the AI chatbot could not fulfill — it required human creativity, taste, and conversation.

IKEA retrained their customer service staff as interior design consultants. They created an entirely new division where human staff work alongside AI tools to provide personalised design recommendations. The human brings aesthetic judgment, empathy, and the ability to understand what a customer really wants. The AI brings product knowledge, inventory awareness, and rapid visualisation.

The result: $1 billion in additional sales attributed to this new AI-augmented interior design service. Not a single customer service job was eliminated — they were elevated. Staff who previously answered routine questions about delivery times are now helping customers design their homes. They report higher job satisfaction, and IKEA reports higher revenue per customer interaction.

This is the model that separates companies with imagination from companies that see AI only as a cost-cutting tool.

Why project managers are running circles around engineers (and what it means for every professional)

One of the most surprising developments in the AI era is happening inside Big Tech — and it flips the traditional power dynamic between product and engineering teams.

Dan Thomasset, a Principal Engineer at Google, posted an observation on LinkedIn in March 2026 that sparked thousands of reactions: "PMs are running circles around SWEs [software engineers] with vibe coding and GenAI prototyping tools, and this is a good thing. The dam has broken for the creatives, and they don't need engineers for the initial phases of their work anymore. Gone are static Figma demos, in with rapid prototyping against production systems."

The implications are enormous. Thomasset described what he calls "the new organisational API" — a fundamentally different way for product and engineering teams to collaborate. Instead of the old model (PM writes a spec, engineers interpret it, vision drifts through multiple rounds of meetings and revisions), the new model eliminates the translation layer entirely. PMs now build working prototypes with AI tools and hand engineers finished, vetted implementations to make scalable and production-ready.

As one commenter put it: "The advantage is not that PMs code better. It is that they removed the translation layer. Before: PM writes spec, engineer interprets it, drift accumulates. Now: PM owns the spec AND the first prototype, so the interpretation gap does not exist."

Pete Simard, a verified tech commentator, added: "PMs have always been translating between humans and engineers — now they can skip the engineer for the prototype. The bottleneck was never the idea, it was waiting three sprints to see if it was even worth building."

Why are PMs thriving while some engineers feel threatened? The answer reveals something important about the AI era: the skills that matter most are not purely technical. PMs succeed with AI because they are good communicators, systemic thinkers, and relationship builders. They understand user needs, they can articulate requirements clearly, and they can navigate organisational complexity. These are exactly the skills that AI tools need from their human operators.

Thomassett confirmed this in a reply: "Communication skills are both the most important part of engineering and the least emphasised in traditional engineering education." A software engineer commenting on the thread agreed: "A major determinant for successfully using AI is the ability to write with precision and clarity. PMs and tech writers tend to do this already."

The skills that AI cannot replace (and why they are your career insurance)

If AI automates routine tasks and amplifies individual capability, what remains uniquely human? The answer is not what most people expect. It is not creativity in the abstract — AI can generate creative content. It is not analysis — AI can process data faster than any human. The skills AI cannot replace are fundamentally relational and contextual.

Stakeholder management and organisational navigation. AI cannot read the room in a tense board meeting. It cannot sense that the CFO is skeptical because of a failed initiative two years ago, or that the VP of Engineering needs to feel ownership before supporting a proposal. Navigating organisational complexity — understanding who has influence, who has concerns, who needs to be aligned before a decision — is a human skill that becomes more valuable as AI handles the tactical work.

Trust-building and relationship development. A 2024 study by Edelman found that 67% of employees say trust in their direct manager is the single biggest factor in their job satisfaction and performance. AI cannot build trust. It cannot have a vulnerable conversation with a struggling team member, celebrate someone's promotion genuinely, or navigate the delicate politics of giving honest feedback to a senior leader. These interactions define careers.

Communication with precision and empathy. The Google thread made this explicit — the professionals who thrive with AI are those who can write clearly, articulate requirements precisely, and communicate across different audiences. This is not a new skill, but AI makes it dramatically more important. When you can translate a business need into a clear prompt and get a working prototype in hours, communication becomes the rate-limiting skill.

Systemic thinking. Understanding how changes in one part of an organisation ripple across others. Seeing the second and third-order effects of a decision. Connecting dots between a customer complaint, a product gap, and a competitive threat. AI is excellent at pattern recognition within data — but humans are still far superior at connecting patterns across domains, politics, and unstructured organisational context.

What Jensen Huang gets right: AI literacy is the new baseline

Jensen Huang has been more explicit than any other major CEO about what AI means for hiring. In multiple interviews throughout 2025, he has repeated the same message: "I would hire the one who is expert in using AI." This is not a prediction about the distant future — it is a statement about hiring decisions happening right now.

The implication is uncomfortable for many professionals: AI literacy is no longer a differentiator. It is becoming a baseline requirement, like knowing how to use email or a spreadsheet. The question is not whether you should learn AI tools — it is how fast you can become proficient.

But Huang's message contains a second, more optimistic insight that often gets overlooked. He does not say AI replaces workers. He says: "AI elevates workers. Every carpenter or plumber could now be an architect." This is the democratisation argument — AI does not just automate, it amplifies. A marketing manager with AI tools can do the analytical work that previously required a data scientist. A project manager with AI can prototype what previously required an engineering team. A salesperson with AI can research and prepare for meetings at a depth that previously required an analyst.

The winners are not the people who know the most about AI — they are the people who combine AI fluency with deep domain expertise and strong professional relationships. A PM who can prototype with AI AND navigate the stakeholder landscape to get that prototype approved and funded will be unstoppable. An account executive who uses AI for research AND maintains genuine relationships with their clients will outperform any AI tool acting alone.

This is where the IKEA lesson meets the Huang principle. IKEA did not just give their staff AI tools — they combined AI with the human skills their staff already had (empathy, design sense, customer understanding) to create something neither humans nor AI could do alone.

Industry by industry: how AI is reshaping specific careers right now

The AI transformation is not theoretical — it is happening across every industry in 2025-2026. But the pattern is remarkably consistent: AI automates the routine, humans handle the relational and the strategic. Here is what is actually happening in the industries that employ the most knowledge workers.

Financial services. JPMorgan's COiN platform reviews commercial loan agreements in seconds — a task that previously took lawyers 360,000 hours per year. But the bank has not eliminated its legal team. Instead, those lawyers now focus on complex negotiations, client advisory, and regulatory strategy. Goldman Sachs reports that AI has reduced the time analysts spend on data gathering by 75%, freeing them to spend more time on client relationships and strategic recommendations. The most valuable skill in finance is no longer spreadsheet mastery — it is the ability to interpret AI-generated analysis and communicate it persuasively to clients.

Healthcare. AI diagnostic tools now match or exceed radiologist accuracy for certain conditions. But the American College of Radiology found that rather than reducing radiologist employment, AI has increased it — because AI-assisted radiologists can process more cases and spend more time on complex diagnoses that require clinical judgment. The World Health Organisation projects a healthcare worker shortage of 10 million by 2030 despite AI adoption, because AI is expanding the scope of care, not replacing caregivers.

Consulting. McKinsey, BCG, and Bain have all deployed AI tools for data analysis, market research, and presentation drafting. McKinsey reports that AI has reduced the time consultants spend on data gathering by 40%. The result: consultants spend more time on client relationships, stakeholder alignment, and strategic advisory — the work that justifies premium fees. Junior consultants who use AI effectively now produce senior-level analysis, but the relationship skills that win and retain clients remain fundamentally human.

Technology. The Google PM story is just the most visible example. Across Big Tech, non-technical roles are gaining influence because AI tools give them execution capability that previously required engineering teams. Product designers prototype interactive applications. Marketing managers run sophisticated data analyses. Operations teams build custom dashboards. The bottleneck has shifted from "who can build it" to "who knows what should be built and can align the organisation to support it."

Education. Teachers report that AI handles administrative tasks (grading routine assignments, creating lesson plans, tracking student progress) in a fraction of the time. A 2025 UNESCO study found that teachers using AI tools spend 30% more time on direct student interaction — mentoring, personalised guidance, and relationship building with students and parents. AI has not made teachers redundant. It has made the human parts of teaching more central.

The data on AI and jobs: what the research actually shows

The media narrative — "AI will take your job" — is not supported by the data. Here is what the most rigorous research actually shows.

MIT study (2024): "The Work of the Future." Researchers analysed 1,200 occupations and found that fewer than 23% of worker wages are associated with tasks that could be cost-effectively automated by AI. The key word is "cost-effectively" — even where AI can theoretically do a task, the economics of deployment, training, and integration often favour augmenting human workers rather than replacing them.

McKinsey Global Institute (2024). Their analysis of 800 occupations concluded that approximately 30% of work activities across all occupations could be automated by 2030 — but this translates to changes in how jobs are performed, not wholesale elimination. They project that AI will create demand for 20-50 million new jobs globally in AI development, deployment, and oversight alone.

World Economic Forum Future of Jobs Report (2025). The most comprehensive survey — covering 1,000+ employers representing 14 million workers — projects that AI and automation will create 170 million new jobs while displacing 92 million by 2030. A net gain of 78 million jobs. The report emphasises that the displacement is concentrated in routine data-processing and administrative roles, while the creation is concentrated in technology, sustainability, relationship management, and creative roles.

LinkedIn Workforce Report (2025). Job postings requiring AI skills grew 2,000% between 2022 and 2025, but overall job postings did not decline. The fastest-growing job categories are roles that combine AI proficiency with human skills: AI-augmented customer success, AI-assisted healthcare, AI-powered financial advisory.

The IKEA lesson, quantified. IKEA's experience is not an outlier. A 2025 Harvard Business School study of 200+ companies that deployed AI found that 62% created new roles or departments as a direct result of AI implementation, while only 18% reduced net headcount. The most common outcome (43% of companies) was simultaneous automation of routine tasks AND creation of new human-centred roles — exactly the IKEA pattern.

The data is consistent: AI changes the composition of work, not the quantity. The professionals who adapt — by learning AI tools and doubling down on human skills — are not just surviving. They are moving into higher-value, more satisfying work.

What leaders get wrong about AI implementation (and what they should do instead)

The IKEA story is not just inspiring — it is instructive. It reveals the fundamental mistake most companies make with AI, and the alternative that creates value.

The mistake: treating AI as a cost-cutting tool. Most companies adopt AI with a simple goal: do the same work with fewer people. This is the industrial-era playbook applied to information-era technology, and it systematically destroys value. When you fire your customer service team after deploying a chatbot, you lose the institutional knowledge, customer relationships, and human judgment that no AI can replicate. You save salary costs and lose the capacity for innovation.

The IKEA approach: treating AI as a capability amplifier. IKEA's leadership asked a different question. Not "how can we do the same with less?" but "what new things can we do now that AI handles the routine?" That question led them to a $1 billion interior design division. The same customer service staff — with the same salaries — now generate dramatically more revenue per interaction because AI freed them to do higher-value work.

Jensen Huang's framework is the same: "Firms with imagination do more with more, not less." NVIDIA could automate many internal functions with AI. Instead, they use AI to amplify their workforce's capabilities and keep hiring. The result is a company that has grown from $17 billion in revenue (2021) to over $130 billion (2025) — with a larger workforce, not a smaller one.

What this means for you as a professional. If your company is deploying AI primarily to cut headcount, that is a signal about your leadership's imagination, not about your value. The companies that will win the next decade are those that use AI to unlock new capabilities, new markets, and new value from their existing workforce. If you are in a company with imaginative leadership, invest deeply — you are about to become much more valuable. If you are in a company that sees AI only as a cost play, start building your relationship network outside the organisation, because you will want options.

If you are a leader, the IKEA lesson is clear: before you automate and reduce, look at what the remaining human work reveals about unmet customer needs. Your AI deployment should not just reduce costs — it should surface insights that create new revenue. The companies that understand this will compound their advantage. The ones that do not will save 20% on labour costs and miss the $1 billion opportunity sitting in their data.

The future was written two years ago: from knowledge economy to curation economy

In March 2024 — two full years before the Google PM thread and the IKEA billion-dollar story became common knowledge — Sorin Ciornei wrote about this exact shift in "The Future is Now: AI's Transformation of Art, Workforce, and Media Consumption" on thereach.ai.

The central argument was prescient: we are moving from a knowledge economy, where value comes from what you know, to a curation economy, where value comes from how you select, combine, and apply information. AI has made knowledge abundant and nearly free. The scarce resource is no longer information — it is judgment about which information matters, and the relationships to act on that judgment.

This is exactly what we see playing out. The Google PMs are not valued because they know how to code — AI gives them that capability. They are valued because they know which product to build, which stakeholders to align, and how to communicate the vision clearly enough that both AI tools and engineering teams can execute on it.

The IKEA customer service staff are not valued because they can answer product questions — AI handles that. They are valued because they can understand a customer's taste, build rapport, and translate vague desires ("I want my living room to feel cosy but modern") into concrete recommendations.

For every professional reading this, the implication is clear: your career advantage is not what you know. It is who you know, how you manage those relationships, and how well you can combine AI capabilities with human judgment. The professionals who systematically track their relationships — who they have met, what was discussed, what matters to each stakeholder, when to follow up — will have an enormous edge over those who rely on memory and good intentions.

This is not a theoretical argument. It is the reason tools like Orvo exist: to give professionals a systematic way to manage the relationship intelligence that AI cannot automate. Your AI tools make you faster and more capable. Your relationship management system ensures that capability is directed at the right people, at the right time, for the right reasons.