Article

AI moves the effort,
it doesn't remove it.

Feb 22, 2026 by Fleur Augustinus

I am a User Experience Designer;
this role requires me to question all the time. I’ve been asking questions since childhood and this has not stopped during my career, it’s an important factor in my life and work, to take the whole ecosystem into consideration before I act.

So my question to the CEO when he asked me to create a workflow using AI;
What is it that you actually want to achieve?

His answer:
"Speed up the workflow to create assets for our sales pipeline, reducing the capacity needed while replacing it with AI tools."

Hold your horses!

I am one of those people who is actually sceptical about what AI can actually do for you. Don't get me wrong; I love new technology, I love smart people inventing yet another tool that promises to make your life easier. Emphasis on promises.

Does AI actually make our life easier?
Or does it just move the effort somewhere else?

AI doesn't give you time back.
It changes where the time goes.

Let's take a step back and look at history. (You are free to skip this boring part)

The irony of the industrial revolution:
it promised to take part of our work away. We ended up working just as much.
Then the internet… it gave us an abundance of information, peaking with the social media era and its endless doomscrolling, information in split seconds, always on, always more.

The world is going so fast right now that we are all collectively burning out trying to keep up with the speed of technology. Which is, genuinely, exhausting. And here we are again. New technology, same promise.

Altho I must confess; I was extremely excited to start the battle: AI vs. me as a designer. All these promises of creating copy, presentations and landing pages in split seconds, while I lean back, eat my breakfast, and by the time I take the second sip of my tea, everything is done and I can go back to bed.

Little side step here: obviously we all can see by now that no one is actually going back to bed. Instead of enjoying the time we supposedly gained…

a moment to breathe, some fresh air, maybe a second cup of tea you actually finish while it's still warm...


we decided to stay inside, behind our screens, in our 90-degree sitting angle (also known as the most unhealthy position for the human body) to create more output. To stay in the race. While being afraid AI is going to make us extinct.

Wait, what?!!

We don't reduce the scope, we don't lower expectations, we don't change working hours(1).

We just raise the bar for what counts as a normal week's output. Instead of "I created one landing page this week", it becomes "I created five landing pages, three email flows and a sales deck, because AI helped." And now that is becoming the standard. Creating endless content for endless scrolling. If you don't have design, development, or critical thinking muscles, it is scarily easy to just trust *the machine*, because it sounds confident and looks "professional enough". That is how low-quality content quietly slides into your sales decks, your website, your client communication.

…and then into someone else's training data. AI doesn’t eliminate thinking either. It shifts it. We move from creating to supervising.

Is this actually correct?
Is this on-brand?
How do I write a prompt so it actually does what I want it to do?


We exhaust ourselves in a race where we have the ability to set the rules, but somehow make those rules really hard, no chance for us to win. #Irony

There is something we don't talk about enough: the friction that used to exist was also doing a job.
When building something took time, budget, and a developer; you thought twice about whether it was actually worth building. That hesitation was not inefficiency. It was a filter. AI removes the filter. Which means we are now building a lot of things that probably should have stayed ideas.

"but think about the productivity!"

If we don’t consciously decide what we are going to do LESS of when we bring in AI, “productivity” is just a fancy word for “longer to‑do lists and higher expectations in the same amount of time”.

AI works best when you design the ecosystem first.

This is the part where my UX brain activates!
AI is not a magic button. It is another tool in our workflow.  If you just start and hope for the best, you get exactly what you would expect if you dropped a very confident, slightly chaotic intern into your team with zero onboarding.

Before asking "which AI tool should we use?",

I want to ask:
What is the actual outcome we care about?
What is the current workflow to get there?
Which tools are we already using?
Where are the real bottlenecks?


And then look at which tasks are repetitive and structured enough to hand off to AI. Like drafting a project plan for example, or do research on an industry or segment (do ask for the source and double check it), Drafting an Ideal Customer Persona profile, Outlining a sales deck or landing page, Writing a rough first draft of copy in a specific tone of voice…These are tasks where AI gives you a starting point, not the final outcome.

At this stage no client-sensitive data is shared, no design system inter-pretended wrong, and no designers, developer or copywriter harmed. After that, things get trickier…I’ve tried to beautify slides with Gemini that turned carefully structured decks into one big image with non-editable text. I have seen AI-generate templates completely ignore the brand guidelines provided in the prompt, or creating copy literally rewriting history with wrong dates included (4).

At that point I am not saving time. I am cleaning up someone else's mess. If I can build a page faster in Webflow using my existing library than I can by writing a ten-paragraph prompt, that is not a win for AI (3). That is a win for good old-fashioned human UX Designer: understanding your system, your constraints, and your users (including your your team).

In a team where workflows, tools and/or a design system is in place (I don’t mean "we use blue and this sans-serif font", but a real visual language that a design team thought carefully about.)

Simple no‑code automations can already deliver the real win. In most cases it is much more powerful to connect the tools you already have than to drop a shiny AI on top of everything and call it innovation. The question isn't "how do we get AI into the workflow?"
The question is: "what is the workflow, and where does AI actually make that journey smoother for humans, instead of just more complicated and time consuming?"

It is about the whole journey, not only the destination.

Let's talk about the elephant in the server room.

There is a lot of money flowing into AI right now (6). Specific type of companies are very excited about you integrating it into your business. These are often not the same people that sit behind laptops trying to keep up with the new expectations this unlocks. I can't help but wonder:is this a gain for me, my team, and our customers? Or is it mostly a gain for another company that does not care so much about values, but does care enormously about dollar signs? If AI mainly helps organisations ship more content faster and push people to do more, while the humans in that system feel more overwhelmed, more replaceable, and more anxious about keeping up … then maybe we are not optimising for the right things?

none of this is free for the planet either. Every "just ask AI" moment sits on top of data centres swallowing huge amounts of energy and water (5). If we are going to burn that much power, I would genuinely prefer it not be primarily to generate another slightly different LinkedIn carousel about 10 ways to triple your growth.

As a UX designer, my job is to care about humans in the system. Not just "the end user", but the people inside the workflow; The designers, the developers, the marketers, the copy writers, and the clients on the receiving end of the final output. So when the CEO asked "can you create a workflow with AI?", I don't just hear "speed up the pipeline."

I hear:
Where will this add friction?
Who will clean up after the AI?
Which skills get quietly devalued?

Because if we don't ask those questions, we will keep building systems where expectations keep going up, attention keeps getting sliced thinner, content keeps multiplying, and our actual wellbeing quietly suffers.
And all of that will be called progress.

I don't want to design for that version of the future.
I want to design ecosystems where AI helps humans breathe.

written by me, the human - structure by Claude and

If you're curious...

...and if you want to go beyond my rant,
look at some research snippets and analysis:

1. AI and working hours
There’s emerging evidence that while AI tools can increase task-level productivity, they often lead to extended workdays and more output expectations instead of more rest.

Doesn’t Reduce Work, It Intensifies It - Harvard business review
2. Digitalisation, overload and burnout
Studies on digitalisation and work consistently link rapid tech change, constant connectivity, and tool overload with role overload, stress, and burnout.
3. Tool overload and “digital debt”
Workplace surveys and reports from large tech companies show that employees spend a big part of their day context-switching between tools and dealing with notifications, which hurts focus and perceived productivity, even when “productivity tools” are meant to help.
4. AI hallucinations and overconfidence
Technical and UX research on large language models documents how often they produce confident but wrong answers, especially in specialised or under-specified contexts. This is why human supervision and domain expertise remain critical.
5. Environmental impact of AI
- Not all AI use is equal; big training runs are far more intensive than individual chats.
- The net climate impact depends on energy mix (fossil vs renewable), efficiency improvements, and what AI is used for (e.g., optimising energy vs generating marketing copy).

But it’s widely acknowledged that AI has a non‑trivial environmental footprint.
6. Capital flows into AI
Reports from financial institutions and industry analysts show that AI investment is heavily concentrated in a small group of large firms and infrastructure providers, raising questions about who captures most of the economic value compared to everyday workers and smaller organisations.

Looking for a human that understands your customer, your team and your techstack?
Say hello and find out how we can collaborate now, or on future projects.
hello@fleuraugustinus.com