All Articles
Future of Work

AI and the traditional working world as we know it.

My view is that AI will not just change jobs. It will change the shape of work itself. More people will operate like consultants, builders, and entrepreneurs, using AI as leverage to package their skills into services, products, and systems that can be exchanged directly for money.

Apr 1, 2026 11 min read Work and AI
AI and the future of work illustration

For a long time, the standard model of work was simple: people trained for a profession, joined a company, traded time for salary, and slowly climbed through a structure. AI introduces a different possibility. One person with the right knowledge, the right tools, and the right discipline can now do the output that once required a whole team.

That does not mean every company disappears. It means the unit of production may shrink. A skilled person using AI well can research faster, write faster, analyze faster, code faster, market faster, and deliver faster. That changes bargaining power, business formation, and the meaning of being employable.

Main point AI is not only a software trend. It is a pressure on the economic structure of work.

The traditional working model is becoming less stable

The old model was built around coordination cost. Businesses hired layers of people because organizing work, producing documents, handling customers, processing information, and creating outputs took a lot of human effort. AI reduces some of that coordination cost.

Less dependency on large teams A smaller group can now produce proposals, designs, reports, software, support drafts, marketing copy, research summaries, and internal documentation much faster than before.
More output per person The worker who knows how to prompt, review, automate, and orchestrate tools can multiply their own capacity.
Pressure on middle layers Work that mainly involves translating, formatting, routing, summarizing, or rewriting information becomes easier to compress.
Faster movement into self-employment Individuals can package skills directly into services without first needing a big team or high startup capital.

In that environment, the future may belong less to people who fit neatly into a job title and more to people who can use tools to create clear outcomes.

A future of consultants, entrepreneurs, and small AI-powered businesses

I think more people will end up working like independent operators. Some will be solo consultants. Some will run small niche agencies. Some will build micro-software businesses. Some will manage AI-supported service firms with only a handful of staff. The common thread is this: AI becomes the multiplier, and the human becomes the strategist, reviewer, relationship holder, and decision-maker.

Consultants People with domain knowledge will use AI to research, prepare, analyze, draft, and communicate at a level that makes them look like a much larger firm.
Entrepreneurs Small businesses will use AI to reduce admin drag, improve sales follow-up, create content, automate onboarding, and support customers without needing huge headcount.
Specialist operators People with a valuable niche skill will sell that skill more directly, with AI helping them package, deliver, and scale it.
Hybrid workers Many people will still be employed, but they will increasingly be expected to operate like mini-businesses inside the company, bringing more leverage per seat.

This shift could be liberating for some people. It gives more room for ownership, flexibility, and direct value exchange. But it can also be brutal for those whose work depends on repeating processes that AI can increasingly imitate or accelerate.

The rise in unemployment is a real possibility

We should be honest about this part. AI may not remove all jobs, but it can reduce how many people are needed for the same amount of output. That means unemployment pressure can rise, especially in roles where the skill is sophisticated but still programmable.

Skilled but repeatable writing work Drafting, rewriting, formatting, proposal assembly, and first-pass content generation are increasingly compressible.
Structured analysis work Roles based on templated research, spreadsheet interpretation, classification, comparison, and summary production are vulnerable.
Routine coding and implementation work Programmers who only translate clear specs into standard code patterns may face more competition and lower pricing power.
Support and operations processing Ticket routing, repetitive replies, intake capture, document checking, and admin coordination can be heavily automated.

The painful irony is that some of the people most at risk are not unskilled. They are skilled in a way that can be decomposed into patterns. If a skill can be clearly instructed, checked, and repeated, it is increasingly a candidate for automation support or replacement.

What still matters on the human side

The answer is not to become anti-AI. The answer is to move toward work where judgment, trust, taste, accountability, negotiation, leadership, and real-world ownership matter more. The more a role depends on ambiguity, consequence, human context, and business tradeoffs, the more durable it becomes.

Judgment AI can generate options. Humans still need to decide what is right, safe, worth doing, and aligned with reality.
Trust People buy from people they trust. Relationships still close deals, calm problems, and create loyalty.
Responsibility When stakes are high, someone must stand behind the final decision. AI does not carry that burden. Humans do.
Creative direction AI can produce many outputs, but strong human direction is what makes output useful, differentiated, and commercially effective.

How we can blend with this change instead of getting crushed by it

We need practical adaptation, not panic. The goal is not for everyone to become a machine-learning engineer. The goal is for more people to become capable operators in a world where software and AI are part of ordinary work.

Learn to work with AI daily Use it for drafting, structuring, exploring, planning, summarizing, and prototyping, but always review the output and sharpen your own thinking.
Build a skill plus a system A skill alone is no longer enough. Pair your expertise with repeatable workflows, templates, automations, and tools.
Create direct market value Learn how your work connects to revenue, savings, speed, risk reduction, or customer satisfaction. General effort is less defensible than measurable value.
Own small assets Build a client list, a product, a niche audience, a reusable framework, or a portfolio of case studies. Assets give stability when roles become unstable.
Move closer to the problem The closer you are to real business pain, the harder you are to replace with generic automation.
Stay multidisciplinary The strongest people in the next era may be those who can combine domain knowledge, communication, systems thinking, and technical fluency.

Some ideas for society, businesses, and individuals

If this transition accelerates, the response cannot only be personal. Businesses, educators, and communities will need to adjust as well.

Businesses should retrain before replacing Teach people how to use AI to increase leverage before deciding they are obsolete.
Schools should teach digital reasoning earlier Not only office software, but logic, systems, automation, data thinking, and how computers follow instructions.
Communities should support micro-entrepreneurship More people may need to earn from small service businesses, digital consulting, niche products, and local problem-solving.
Workers should build portfolios, not only CVs In an AI-heavy economy, proof of outcomes matters more than a list of responsibilities.
Governments and institutions should monitor transition pain If productivity rises while employment compresses, serious social and economic planning will be needed.

Everybody should start learning a new language: computer

I do not mean everybody must become a full-time software developer. I mean everybody should begin to understand how computers think, how instructions work, how data flows, how automation is chained, and how software can be directed to do useful work.

In the same way that reading and writing changed a person's power in the industrial and information eras, computational literacy may shape a person's power in the AI era.

Learn prompting properly Clear inputs produce better outputs. Learn structure, context, examples, constraints, and iteration.
Learn basic logic Understand conditions, steps, variables, inputs, outputs, and error states. This is the grammar of automation.
Learn simple scripting or no-code automation Even basic exposure to code, spreadsheets, APIs, or workflow tools changes how you see work.
Learn data hygiene Messy data leads to weak AI. Clean information, naming discipline, and source quality matter.
Learn verification habits AI is powerful, but it can still be confidently wrong. Checking outputs is part of literacy.
Learn system thinking Stop seeing tasks in isolation. Learn to see the flow between people, tools, information, approvals, and outcomes.
Plain version The more you understand how to instruct software, the more useful AI becomes to you instead of dangerous to you.

The future will reward leverage, clarity, and adaptability

I believe AI will push many people away from depending only on traditional employment structures and toward more independent, outcome-driven work. Some of that shift will feel empowering. Some of it will feel harsh. Both can be true at once.

The people who do well will likely be those who combine human judgment with technical fluency, turn their skills into systems, and learn how to use AI as a tool rather than a threat. The next language many people need to learn is not only English, Zulu, or French. It is computer.