AI and the Workplace: What's Changing?

In recent weeks, Anthropic released Opus 4.7  and Claude Design, an AI feature that helps users create designs, interactive prototypes, and slide decks through conversational prompts.

This new update has the whole design community in a frenzy. 

Even if you hadn't thought about it before, news like this jolts you to reality and forces the million-dollar question to your face: Will AI actually take your job?

Amidst the hot takes and views, one perspective stands out, a big part of a designer's job, anyone's job really, is helping clients figure out what they actually want. Thinking through the problem, providing a solution and communicating the value. 

What This Means For The Workplace

Let's be honest, AI design is decent at best. The ideation process involving context, intent, emotion, and direction is still a human thing. To a large extent, most of computing still runs on one rule that hasn't changed: garbage in, garbage out. 

It takes real design thinking, an understanding of colour, typography, layout, and other design principles to craft a prompt that produces something meaningful. Even after you get a result, you still have to audit, refine, and create something that’s up to standard. This human element isn’t something that just disappears.  

The thing most people miss when they argue about whether AI will take jobs is that the shift is less about humans being replaced and more about redefining how work is done and restructuring the workplace around judgment rather than execution.

For most of history, value came from your ability to do stuff: write, design, code, analyse. That layer is getting cheaper and faster.

So value is migrating from execution-based tasks to decision and direction, what should be done, what good output looks like, and the right course of action. All of which require critical thinking.

The shift can be viewed through three lenses:

Execution

This is where AI thrives currently. Drafting content, generating designs, summarising information, and writing basic code. This layer is becoming faster and less human-dependent. 

Direction

This is a step further than execution and is still a human element of work. Here, people define problems, shape outputs, and iterate toward quality.

This is where designers become creative directors, writers become strategists, and analysts become decision-makers. AI handles the execution, and we determine the direction.

Judgment

This sits at the top. Think: taste, intuition, cultural awareness. Basically, the ability to know what actually matters and why. AI makes the work faster, easier, and more efficient, but we humans define what “efficient” and “good” actually mean.

We are the judges of the quality of AI’s execution. AI doesn't care, prioritise, or understand stakes the way we do. Not yet, anyway.

The Half-truth

There’s a dangerous belief being peddled. Many people believe that AI will just make everyone more productive. Unfortunately, that’s only half true because it doesn’t stop there.

AI will not only increase output but also increase competition. Think about it, if everyone can produce more using AI, average, good-enough quality floods the workplace, and it becomes harder to stand out. Mediocrity gets churned out faster.

Think about how quickly people call out AI-generated content. We can tell when something is off, when it's generic, when it lacks depth, when it just doesn't connect. That instinct is the differentiator.

The workplace needs people whose value doesn’t solely rely on what they can produce, but rather on strategic thinking, taste, clear direction, and communication.

The real threat is to roles that are purely execution-based, repetitive, predictable, and measured by speed and volume. With the current trend, one person with AI can now do the work of many. 

For example, generic content writing like product descriptions can easily be outsourced to AI because most product descriptions follow predictable formats, repeat similar structures, and are written for scale, not requiring deep originality.

The Other Side Of The Story

While certain tasks are shrinking, the tech landscape is expanding, creating roles that didn't exist before.

AI trainers. Prompt engineers. AI auditors. Ethicists working inside tech companies. Specialists who help organisations figure out how to actually implement these tools without breaking their workflows or their culture.

What these roles have in common is that they require someone who understands both the tool and the craft it's being applied to. A healthcare AI auditor who doesn't understand medicine would not be effective. A content AI trainer who can't recognise good writing won't produce good outputs. These new roles are built on the foundation of the old skills.

So, will AI take your job?

Probably parts of it: the parts that were always the most repetitive, the most mechanical, the least human.

But the thinking, the taste, the judgment calls about what actually matters are still human and if anything, that matters more now than it did before.