Where would we be without AI today?
And where are we heading to with it? Should designers learn to use neural networks the way we once mastered Figma? Is this the new entry barrier to the profession?
Alex Perminov, creative web & VR designer at JetStyle, shares his take on this. Read, agree, argue, and share your own experience in the comments.
Which AI tools do you use most often in your design work?
Alex: For me and many folks on our team, the core tool is Midjourney. It’s not perfect, but it can pick up any style and works great for internal PR visuals, presale concepts, and illustrations. When Midjourney struggles, I switch to Stable Diffusion. It’s fussier, but it gets the job done.
And don’t underestimate ChatGPT as an image tool. It’s weak at aesthetics, but surprisingly good at following complex prompts about composition or object placement. It’s great for mockups: just ask it to merge an object from one image into another.
However, I’d also add that Google recently released its Nano Banana, which is similar to GPT but even better. We now use it for the same tasks, but it’s faster and more efficient. What used to be handled by GPT for tasks like composition and object placement is now done seamlessly with Nano Banana, saving us both time and effort.
For narrow tasks, we use Recraft or Krea. Video generators are changing fast; recently I’ve tried Kling, Luma, and Hailuo Minimax.
Can AI fully replace a junior designer?
Alex: Not if that junior knows at least a bit of AI. It’s still a tool. Sometimes you spend hours generating, sorting, and refining results.
AI automates workflows, but not to the level where an art director presses one button and walks away to grab a coffee.
What’s the main benefit of using AI in your daily work?
Alex: It helped me grow and try new things. I started as a graphic designer, then shifted more into interfaces. Without AI, all our beauty shots and concepts would be done by concept artists only.
With neural networks, I can handle it myself, even without being a “true” illustrator. It saves time, but also creates more tasks.
Where does AI save you the most time?
Alex: Internal stuff: PR and HR visuals, social media, event posters.
Neural networks are amazing for creative ideas and unusual visual solutions. In commercial work, they’re perfect for concepting and placeholder content when the client doesn’t yet have a clear visual vision.
Should designers learn AI, or just use it when needed?
Alex: They absolutely should. Even if it doesn’t feel essential now, it soon will. Better start early.
I’d also add that, beyond just learning to use neural networks, it would be a big bonus if designers delve more into the technical side of how they work. This deeper understanding will help them better grasp which tasks neural networks can handle and which ones they can’t, and why.
What tools do you use for unusual tasks, such as storytelling, motion, concepting?
Alex: For narratives and brainstorming: GPT, Grok, DeepSeek. Any language model works.
For motion design: Kling, Luma, Hailuo Minimax (the list changes all the time).
And there are niche tools like SkyBox AI for generating 360° panoramas; it’s super useful in 3D and VR projects.
Have AI tools changed the requirements for designers?
Alex: I haven’t interviewed anyone myself, but I guess some employers (not ours!) see AI as a magic bullet. They expect faster results and lower rates. It’s not entirely true, but this idea is out there.
How do you keep stylistic consistency with AI-generated visuals?
Alex: It’s getting easier. AI easily picks up a reference style and stays consistent. Many tools now let you train mini-models on your own reference set.
Do you use AI beyond motion, images, and sound?
Alex: I use GPT for video scriptwriting, and I end up learning random stuff like what a “gantry crane” is while working on a logistics project.
It’s also great for coding micro-scripts: for example, we recently added custom popups to a Webflow site via GPT-generated code.
Will AI replace designers sooner than we think?
Alex: Who knows. Probably not. The market’s unstable, but all the AI panic and “mass layoffs” headlines are totally overblown.
Have you seen any specialists already replaced by AI?
Alex: Nope. Everyone’s just adapted. Some keep expanding their skillset, others use AI occasionally. But people with real skills are still in demand.
So, is using AI cheating or a sign of adaptability?
Alex: 100% adaptability.
Cheating is when something makes your job effortless. AI doesn’t. You’ll swear, rage, and smash your keyboard while learning it, but you’ll come out stronger.
To sum it up
AI isn’t a cheat code, it’s a stress test for adaptability.
Today, working with neural networks is as essential for designers as learning Figma once was.
Manual skills matter less at the start, but structured thinking and curiosity still make all the difference.
Want to discuss how AI can speed up your creative process? Reach us at orders@jet.style.
