What AI Is Actually Doing to Jobs: A Builder's Perspective

52% of employed workers say they're worried AI will replace their jobs. I understand the anxiety. But I've spent the past six months building AI-integrated systems for real organisations, and the picture on the ground is more nuanced than the headlines suggest.

Here's what's actually happening.

Jobs are changing, not disappearing

The World Economic Forum's 2026 analysis found that after ChatGPT's release, job postings for structured, repetitive roles decreased by 13%. But employer demand for analytical, technical, and creative roles grew by 20%.

That's not "AI is killing jobs." That's "AI is shifting what jobs look like."

I've seen this play out in real time. At a sports organisation I worked with, nobody lost their job when I built AI-integrated technology platforms. What changed was how they spent their time. The marketing coordinator stopped manually formatting newsletter content (AI does a first draft now) and started spending that time on strategy and stakeholder relationships. The operations team stopped manually processing registration forms and started focusing on event planning and community engagement.

The work became more interesting, not less. The tedious parts got automated. The meaningful parts got more attention.

The AI skills premium is real

PwC's 2025 Global AI Jobs Barometer found that workers with AI skills command wage premiums up to 56% higher than their peers. That's not a marginal advantage. It's a career-defining difference.

But "AI skills" is misunderstood. It doesn't mean becoming a machine learning engineer. It means:

• Knowing how to use AI tools effectively in your existing role
• Understanding what AI is good at and what it's not good at
• Being able to evaluate AI output critically
• Knowing how to structure requests (prompts) to get useful results
• Understanding the ethical and practical limits of AI-generated content

A marketing manager who can use AI to draft and iterate campaign copy three times faster than before is more valuable than one who can't. An HR professional who can process assessment data with AI assistance and deliver reports in days instead of weeks serves more clients at higher quality. A financial analyst who can load a full year of reports into an AI context and get cross-referenced insights in minutes is doing work that previously required a team.

These aren't technical skills. They're professional skills augmented by new tools. And they're increasingly the baseline expectation, not a differentiator.

What I've watched AI transform

Here are specific examples from my own work over the past six months:

Software development. I'm a single developer who delivered what would traditionally require a team of 4-6 people. AI didn't replace developers - it multiplied my output. The demand for developers hasn't decreased. But the expectation of what one developer can deliver has increased dramatically.

Assessment and reporting. I built Scorafy to automate the manual report-writing that coaches, educators, and HR professionals spend hours on. The practitioners who use it don't do less work - they do better work. They spend less time on data processing and more time on the conversations and relationships that actually drive results.

Content creation. Every organisation I've worked with now uses AI for first-draft content creation. Blog posts, emails, internal communications, documentation. The writing still needs human review and refinement. But the time from "blank page" to "solid draft" dropped from hours to minutes.

Data analysis. Loading datasets into AI contexts for pattern recognition, anomaly detection, and summary generation. Work that would take an analyst a full day produces results in minutes. The analyst's job shifts from "process the data" to "interpret the results and make recommendations."

The three categories of jobs

Based on what I've seen, jobs are falling into three categories:

Category 1: Transformed by AI. These are roles where AI handles the routine components and humans focus on judgment, creativity, and relationships. Most knowledge work falls here. The job title stays the same. The job description changes significantly. Examples: marketing, HR, finance, consulting, education, software development.

Category 2: Reduced by AI. Roles that were primarily about processing structured information in predictable ways. Data entry, basic bookkeeping, routine report compilation, template-based document creation. These roles are genuinely shrinking. The work still exists, but it requires fewer people.

Category 3: Created by AI. New roles that didn't exist three years ago. AI strategy consultants. Prompt engineers. AI integration specialists. AI trainers and coaches who help organisations adopt these tools. AI ethics and governance professionals. This category is growing rapidly.

The most important category is the first one. Most workers won't lose their jobs to AI. But their jobs will change, and the ones who adapt will be significantly more productive (and more valuable) than those who don't.

What individuals should do

If you're worried about AI's impact on your career, here's my practical advice:

1. Start using AI tools now. Don't wait for your employer to provide training. Sign up for Claude, ChatGPT, or another AI assistant and start using it for real work tasks. Draft emails. Analyse spreadsheets. Generate reports. Build familiarity through practice.

2. Focus on what AI can't do. Relationship building, creative judgment, ethical reasoning, domain expertise, leadership, communication. These are the skills that become more valuable as AI handles more of the routine work. Invest in them deliberately.

3. Become the person who teaches others. In every organisation, there's a gap between the early adopters and everyone else. Bridge that gap. Help your colleagues use AI effectively. That makes you indispensable.

4. Think about AI as amplification, not replacement. The best analogy is the calculator. When calculators became standard, accountants didn't disappear. Accounting became more sophisticated because the tedious arithmetic was handled by a machine. The same thing is happening now, at a much larger scale.

What business leaders should do

If you're leading an organisation:

1. Invest in upskilling, not just tools. 53% of organisations are educating their broader workforce to raise AI fluency, according to Deloitte. That's the right priority. Buying an AI tool and hoping people figure it out is a waste of money. Training your team to use AI effectively is the highest-ROI investment you can make right now.

2. Redesign workflows, not just tasks. Don't just ask "which tasks can AI do?" Ask "how should this entire workflow change now that AI is available?" The biggest gains come from rethinking processes, not just automating steps within existing ones.

3. Create space for adaptation. Change is uncomfortable. Give your team time to experiment, make mistakes, and build confidence with AI tools. The organisations that create psychological safety around AI adoption will adapt faster than those that mandate it.

4. Be honest about the transition. Some roles will change significantly. Pretending they won't erodes trust. Have honest conversations about how work is evolving and invest in helping affected employees develop new skills.

The bottom line

AI is not going to replace you. But it is going to change your job. The professionals who treat AI as a capable tool - learning to use it, understanding its limits, and combining it with their own expertise - will be the ones who thrive.

The ones who ignore it will find themselves doing manually what their colleagues do in a fraction of the time. And eventually, that gap becomes impossible to close.

The good news: the learning curve is gentle. Start today. Use AI for something real this week. See what happens. That first step is the most important one.

If your organisation needs help making this transition, get in touch. I work with businesses to build practical AI strategies that deliver results, not just slide decks.