AI makes Skills more valuable, not less.

Chidiadi Oscar — Thu, 30 Apr 2026 12:03:30 +0000

For years, the dominant narrative around AI has been simple: machines are coming for jobs. We’ve heard this framing so often that it’s become background noise, but it rests on a flawed assumption.

The assumption is that jobs are fixed bundles of tasks, and if you automate enough of those tasks, the job itself disappears.

That’s not how work actually functions.

As Jensen Huang recently articulated, a job isn’t defined by the tasks. A job is defined by its purpose. The tasks are just implementation details.

Once you grasp this distinction, everything about the AI-and-work conversation begins to look very different.

Jobs Were Never About Tasks in the First Place

Ask someone what a lawyer does, and they’ll tell you:

they write contracts
they research case law
they prepare arguments
they draft legal documents

Ask a marketer, and you’ll hear about writing copy, running ads, analysing metrics, testing campaigns.

Ask an engineer, and they’ll mention code, debugging, documentation, building features.

This is how we’ve always talked about work. We inventory the activities, we make lists, and for a long time, it’s been very useful in describing what we do.

However, If you zoom out slightly, you realise that none of these things are actually the job. They’re all intermediate steps.

A law firm doesn’t hire lawyers to produce documents. If that were the job, why would they exist now that document templates exist?

The reason law firms exist is because clients need someone to navigate a legal system and protect their interests.

The documents are just evidence of that work. They’re the artifact, not the purpose.

The same applies to marketing.

A company doesn’t hire a marketer to write ads. The ads are the output.

What they’re really paying for is someone who can understand a market and influence people toward a desired outcome.

And with engineering, no one cares about the code itself. They care about the business problem the code solves.

Once you separate activity from purpose, the AI-and-jobs conversation becomes clearer.

AI Does Not Replace Jobs — It Removes Task Friction

AI is objectively good at certain things:

drafting
summarising
retrieving information
generating variations
structuring data
rewriting text

All of these live in the execution layer. They are the how of getting work done.

But there are parts of work AI struggles with:

defining intent
deciding what the goal should be
taking responsibility for outcomes
making judgement calls under ambiguity
being accountable when things go wrong

That’s where human value concentrates.

So what actually happens is this:

AI compresses the execution layer.

It shortens the distance between deciding what needs to happen and producing the output.

This is where people misread the situation.

They see AI doing the work and assume the job is obsolete.

But what actually happens is that the job changes.

The friction that used to consume most of a professional’s time disappears, forcing them to move upstream—closer to the purpose of the role.

Work Is Defined by Purpose

Take the legal profession.

For decades, being a good lawyer meant:

drafting agreements
researching precedent
writing legal memos

These are valuable skills but they are not the job.

The job is:
achieving the best outcome for a client within a legal system.

Now introduce AI.

It can:

draft agreements
retrieve relevant precedent
summarise case law

Suddenly, a lawyer’s day looks different.

Instead of spending hours producing documents, they spend more time:

reviewing outputs
advising clients
negotiating outcomes
making strategic decisions

What changed?

The lawyer didn’t lose their job.

The job got compressed.

Execution was reduced.

Purpose remained.

The Hidden Effect: Judgement Becomes More Valuable

When execution friction disappears, something else becomes the bottleneck:

judgement.

Not the ability to do something—but the ability to decide what should be done.

Judgement involves:

understanding context
identifying constraints
choosing desirable outcomes
making trade-offs

This cannot be easily systematised.

It becomes more valuable as execution becomes cheaper.

When anyone can produce output, the scarcity becomes:

knowing what to produce.

This is why AI does not flatten skill hierarchies.

It reshapes them.

The Shift From Doers to Directors

For most of the 20th century, value came from execution.

The best writer, engineer, or analyst had an advantage because execution was scarce.

AI changes that.

Execution becomes cheap.

So value shifts upward:

from doing → directing
from producing → deciding
from output → outcome design

This introduces new high-value skills:

Problem Framing
The ability to clearly define what needs solving.

System Thinking
Understanding how different parts of a workflow interact.

Decision Design
Structuring choices to produce consistent outcomes.

Context Engineering
Providing the right inputs so systems produce useful outputs.

These are not new skills,they were just hidden before.

Why High-Skill Workers Become More Valuable

A common fear is that AI levels the playing field.

It doesn’t.

AI removes baseline effort, but it amplifies thinking.

A mediocre thinker with AI becomes slightly more productive.

A strong thinker with AI becomes exponentially more effective.

AI amplifies direction—not intelligence.

If you know what you want, AI accelerates you.

If you don’t, it just produces confusion faster.

The Real Bottleneck Has Moved

Before AI:

writing took time
research took time
analysis took time

Execution was the bottleneck.

Now:

drafts take minutes
research is instant
outputs are immediate

The bottleneck moves upstream.

It becomes:

clarity of intent
quality of problem framing
strength of judgement

AI is not inconsistent.

It is extremely consistent with the input it receives.

Good input → good output

Confused input → confused output

Why “AI Will Replace Jobs” Is the Wrong Question

The real question is:

What part of a job is actually the job?

If you remove all tasks, what remains is:

purpose
responsibility
judgement
direction

These are difficult to automate.

Because they are context-dependent.

And context is where humans still dominate.

What This Means for Professionals

If you work in law, marketing, engineering, writing, or analysis, your job is not disappearing.

But it is changing.

You will spend less time:

producing outputs
doing repetitive tasks

And more time:

reviewing work
making decisions
defining goals
structuring systems

This is a shift from producer to director.

And that’s where the real value has always been.

The New Competitive Advantage

In an AI-driven world, the most valuable skills are cognitive:

clarity of thinking
precision in defining goals
structuring ambiguity
guiding systems toward outcomes

These skills were always important.

But now they are visible.

And that visibility creates a new hierarchy.

Conclusion: AI Removes Tasks, Not Purpose

Jensen Huang’s insight is simple:

A job is defined by its purpose, not its tasks.

AI removes the how.

It does not remove the why.

The lawyer still delivers legal outcomes.

The marketer still drives results.

But they no longer spend most of their time on execution.

What remains is the part of work that matters most:

judgement
direction
decision-making

These are the things AI does not replace.

It makes them more valuable.

Why Your AI Outputs Are Weak: A Prompt Design Problem

Chidiadi Oscar — Wed, 15 Apr 2026 10:53:30 +0000

Poor AI output is often blamed on the model. In most cases, it is a prompt design problem.

Many users approach LLMs with a search mindset. They type short, vague queries and expect useful results. That works for search engines. It does not work for generative systems.

Search vs Generation:
Search engines retrieve information based on keywords. LLMs generate responses based on:

structure
context
constraints

This difference is where things start to look difficult, especially when you try to use retrieval systems for AI generation.

If you type:

marketing ideas

A search engine returns articles, videos, and frameworks.

An AI system has to generate an answer from scratch. With no clear direction, the result is usually broad and unfocused.

The issue is not the model,it is the lack of instruction and clarity in the prompt.

LLMs Do Not Understand Language Like Humans:

Another common mistake regular people make is assuming that AI understands meaning. LLMs work by predicting patterns in text, after which they generate responses based on probability, not understanding.

This creates a problem:

fluent output can still be wrong
confident tone can still be misleading

Fluency is not accuracy and as a user, you should always vet and verify every output from LLMs.

Prompt Structure Determines Output Quality:
Unstructured prompts create unclear outputs. When your input is vague, the system has too many possible directions and as a result it defaults to generic responses.

Example:
Unstructured:Explain marketing
Structured: Structured:Explain three practical marketing strategies for a new e-commerce business. Include one example per strategy.

The second prompt works better because it defines:

scope (three strategies)
context (new e-commerce business)
format (explanation + example)

Less ambiguity leads to better results:

Constraint improves the output of every interaction.Constraints are rules that shape the response.
They define:

what to include
how to structure it
how detailed it should be

Without constraints, the output is general. This is because constraints limit the interpretations every LLM has to make when processing user requests, which enables the LLM to provide contextualised output. Without this,the LLM provide broad and general output that does not help the user.

Better Outputs Come From Better Instructions and Clear Prompts:
Switching tools does not fix weak results.The same model can produce very different outputs depending on the prompt.

What matters is:

clarity
specificity
structure These are controlled by the user.

Conclusion:
Weak AI output is not a model problem, it is an instruction and prompting problem.

Clear prompts produce clear results, unclear prompts produce generic ones.

If you want better outputs, improve how you give instructions.

DEV Community: Chidiadi Oscar

AI makes Skills more valuable, not less.

Why Your AI Outputs Are Weak: A Prompt Design Problem