How to Use AI in the Fashion Design Process (Without Losing Your Voice)

Light pale grey two-tone print of a tablet with a stylus, an open sketchbook and a small toy robot on a charcoal blue background

AI image and text tools can speed up some stages of designing a collection, but they can’t do your research, make your decisions or give you clear ownership of the result.

By Shreya S · Last checked 17 November 2026 · 12 min read · Building skills

AIDesignDesign & Creative PracticeWorked exampleChecklist

In short

  • AI is most useful at the edges of the design process, for exploring, varying and presenting ideas, and least useful as a replacement for primary research, drawing and making.
  • Keep AI outputs clearly labelled in your sketchbook and portfolio, and follow your course rules.
  • Copyright in AI-generated images is unsettled. In the US, purely AI-generated material isn’t protected, and the law in the UK and elsewhere is still under review.
In this guide · 15 sections
  1. Where AI fits in a design process
  2. First: check what your course allows
  3. Research: AI as a pointer, not a source
  4. Concept: exploring moods and colour
  5. Development: variations and iterations
  6. Presentation and portfolio
  7. Copyright and ownership: what we know
  8. Ethics and impact
  9. Documenting your use
  10. Writing image prompts that serve your research
  11. AI beyond image generators
  12. Showing AI use well in a portfolio
  13. Questions to ask before using a tool
  14. Common mistakes
  15. Frequently asked questions

Where AI fits in a design process

A typical fashion design process moves through stages like these. AI tools can play a part in some of them, but not all equally.

StageWhat you doWhere AI can helpWhere it can’t replace you
BriefUnderstand the client, customer, season and constraintsRephrase a brief, list questions to askUnderstanding the real customer and context
ResearchPrimary and secondary research: archives, museums, places, people, materialsSuggest themes, keywords and references to look upVisiting, photographing, handling, drawing, interviewing
ConceptChoose a direction and moodGenerate mood variations, colour stories, word associationsChoosing what matters and why
DevelopmentSketches, drapes, collage, toiles, samplesFast variations on a silhouette or print ideaTesting on the body, fabric behaviour, fit
Range buildingEdit the line-up, balance the collectionVisualise line-up options, check repetitionCommercial and creative judgment
TechnicalFlats, specs, costing, constructionDraft descriptions, check spellings of termsAccurate measurements and construction knowledge
PresentationPortfolio, lookbook, pitchLayout ideas, copy drafts, background mock-upsYour story, your voice, the real garments

The pattern is that AI tends to help with speed and variation. The research, judgment and making remain yours.

First: check what your course allows

Before you use any AI tool in a design project, check your module’s rules. Many universities ask assessment designers to state clearly whether AI is permitted. UAL, for example, asks staff to use a red-to-green system for each assessment, where red means AI is not permitted. If AI isn’t allowed in a project, don’t use it. If it is, you may be asked to document how you used it. Our guide How to use AI in your coursework honestly explains how to find and follow the rules.

Research: AI as a pointer, not a source

AI chatbots can help you start your research by suggesting directions you might not have considered. For example:

“I’m designing a menswear collection inspired by 1970s Indian railway workers’ uniforms. Suggest research directions, archives and types of primary research I could do.”

Use the answer as a list of leads, then do the real work:

  • Visit archives, museums and collections. Museum collection websites (the V&A, the Met, the Kyoto Costume Institute and many national museums) let you search objects with dates and descriptions you can cite.
  • Take your own photographs and make your own drawings on location.
  • Handle fabrics, trims and garments. AI can’t tell you how a cloth drapes, weighs or wears.
  • Talk to people. Interviews with makers, wearers and communities give your work depth AI can’t.

Be careful with AI descriptions of historical dress, regional crafts and cultural garments. They can be inaccurate, generic or stereotyped. Check everything against reliable sources, and be thoughtful about cultural appropriation: credit communities and crafts properly, and consider whether and how to collaborate.

Concept: exploring moods and colour

At the concept stage, image generators like Midjourney, Adobe Firefly, DALL·E or Stable Diffusion can produce quick mood images from text prompts. Some designers use them to test the feel of a direction before committing.

Tips if you do this:

  • Generate from your research, not instead of it. Describe your own research findings in the prompt.
  • Combine with your own images. Collage AI outputs with your photographs and drawings rather than relying on them alone.
  • Use them to rule things out. Seeing a direction quickly can help you realise it isn’t right.
  • Label them. Mark every AI image clearly in your sketchbook, with the tool and date.

Colour is an area where AI suggestions can be a useful starting point, but always test colours physically, with dyed swatches, yarns or paint, because screen colour isn’t the same as fabric colour.

Development: variations and iterations

AI image tools can produce many variations on a silhouette, print or detail quickly. This can help you push an idea further. But:

  • AI images don’t show how garments are made. Seams, closures and construction often don’t make sense. A generated garment might be impossible to make.
  • Fabric behaviour is guessed. Drape, weight and movement in AI images are illustrations, not tests.
  • Bodies are often limited. Many generators default to narrow ideas of bodies, skin tones and gender. Push for diversity and be critical of defaults.

Use AI variations as one input among many, and do your real development in sketches, drapes on the stand, toiles and samples.

Worked example: using AI in a print project

A student developing a print collection based on her grandmother’s garden photographs her own plants, draws them in ink and scans the drawings.

She then uses an image generator, which her module allows, to test how one of her motifs might look as a repeat at different scales and in three colourways. She chooses one direction, then redraws and repeats the print herself in Photoshop, because the AI version had distorted the leaf shapes.

In her sketchbook, she labels the AI pages clearly: “Generated with [tool], [date], prompt: …” Her reflection explains what the AI helped with (fast scale tests) and what it didn’t (accurate botanical detail). Her final prints are her own drawings.

This is an invented example to show good practice.

Presentation and portfolio

AI can help with presentation tasks, for example:

  • drafting a first version of a collection statement, which you then rewrite in your own voice
  • suggesting portfolio layouts
  • generating background scenes for mock-ups, clearly labelled

But in a portfolio, employers and tutors want to see you: your research, thinking, drawing, making and decisions. A portfolio full of AI images can raise questions about what you can actually do. If you include AI-assisted work, say so and show your process around it.

The law on AI and copyright is changing fast and differs between countries. This is a general overview, not legal advice.

United States

In January 2025 the US Copyright Office published Part 2 of its report on copyright and AI, on copyrightability. It concluded that copyright protects human authorship, that material generated wholly by AI is not protected, and that prompts alone generally don’t give enough human control to make the user the author of the output. It also said human contributions, such as creative selection, arrangement or modification of AI output, or human-made material included in a work, can be protected.

In March 2025, in Thaler v. Perlmutter, the US Court of Appeals for the D.C. Circuit upheld the Copyright Office’s refusal to register an artwork listed as created autonomously by an AI system, confirming that US copyright law requires a human author.

United Kingdom

UK law includes a rule, in section 9(3) of the Copyright, Designs and Patents Act 1988, that gives protection to “computer-generated works” where there is no human author. How this applies to generative AI is unclear and debated. In December 2024 the UK government opened a consultation on copyright and AI, which included questions on whether to keep, reform or remove that rule. The consultation closed in February 2025, and the government has since published a report and impact assessment on the use of copyright works in AI development. Policy is still developing, so check for the latest position.

In November 2025 the High Court decided Getty Images v Stability AI, the UK’s first major judgment on generative AI and copyright. The court dismissed Getty’s secondary copyright infringement claim, finding that the Stable Diffusion model was not an “infringing copy” because it did not store or reproduce Getty’s images, and found only limited trade mark infringement relating to watermarks appearing in some outputs. The case was narrow, and many questions about training and outputs remain open.

What this means for you

  • Don’t assume you own AI-generated images. In the US, purely AI-generated material isn’t protected by copyright, and the position elsewhere is uncertain.
  • Your own drawings, photographs and garments are much more clearly yours. This is another reason to keep AI as a tool around your own work, not a replacement for it.
  • Read the tool’s terms of service. They set out what you can do with outputs, including commercial use.
  • Be careful prompting with other designers’ names or protected work. Asking for something “in the style of” a living designer, or copying a brand’s logo or prints, can raise ethical and legal issues and can look like copying to a tutor or employer.
  • Brands may have their own AI policies. If you’re on an internship or freelancing, ask before using AI tools on client work, and never upload confidential designs.

Ethics and impact

Beyond rules and law, it’s worth thinking about:

  • Artists and training data. Many artists object to their work being used to train image generators without consent or payment. Some tools say they train on licensed or public-domain material; check what each tool says about its training data.
  • Representation. Generators can reproduce narrow beauty standards and stereotypes. Be critical, and don’t let defaults shape your casting or styling.
  • Environmental cost. Training and running AI models uses energy and water. That’s worth weighing, especially in a sustainability-focused project.
  • Craft and labour. Consider how AI might affect pattern cutters, illustrators, photographers and models, and whether your use supports or undercuts them.

Documenting your use

If you use AI in a design project, keep a record:

  • tool and version
  • date
  • prompt
  • the output (screenshot or file)
  • what you did next with it

A short reflection on how AI helped and where it fell short is also useful, and often required where AI is part of the brief.

Writing image prompts that serve your research

If your course allows image generators, the quality of what you put in shapes what you get out. A prompt built from your own research keeps the output connected to your project.

Weak promptStronger prompt built from research
“Cool futuristic menswear”“Menswear outfit inspired by 1970s railway workers’ uniforms: heavy cotton drill, patch pockets, faded indigo and rust, utility details, photographed on a station platform in flat morning light”
“Pretty floral dress”“Midi dress with a large-scale print of hand-inked garden roses and ivy, loose bias-cut silhouette, muted sage and cream palette, studio shot on a plain background”
“Sustainable fashion look”“Outfit made from visibly mended knitwear, darned elbows in contrasting yarn, layered over a patched denim shirt, natural light, close-up on the repairs”

Notice that the stronger prompts describe your research: materials, references, colours and context. They don’t name living designers.

Iterate and record

  • Change one thing at a time (silhouette, colour, fabric, setting) so you can see what each change does.
  • Save each prompt and output with the date.
  • Note which outputs you rejected and why. Rejections are evidence of your judgment.

AI beyond image generators

AI features are also appearing inside tools designers already use:

  • Adobe applications such as Photoshop and Illustrator include generative features. Adobe says it trains its Firefly models only on data it has the rights to use, and that its first models were trained on Adobe Stock images, openly licensed content and public-domain content whose copyright has expired. Check Adobe’s current documentation for details.
  • 3D design tools used in fashion, such as CLO 3D and Browzwear, simulate fabric and fit on avatars. They’re not generative AI in the same sense, but they also change how designers sample and present work.
  • Text tools can help draft collection statements, product descriptions and grant applications, which you should always rewrite in your own voice.

Your course may teach or provide specific tools, so check what’s available and permitted.

Showing AI use well in a portfolio

If you include AI-assisted work in a portfolio, presentation matters:

  • Label it on the page: “AI-assisted concept image, [tool], [date]”.
  • Show the before and after: your research, the AI exploration, and your own development drawings or toiles that followed.
  • Keep it proportionate: AI exploration should be a small part of a project, unless the project is specifically about AI.
  • Be ready to talk about it: interviewers may ask how you used AI, and why. A thoughtful answer shows judgment.

Worked example: a portfolio page layout

A student’s portfolio spread for a knitwear project shows, left to right:

  1. Research: photographs of fishermen’s ganseys from a local museum visit, with the museum credited.
  2. AI exploration (labelled): six generated colourway tests, with the prompt in small text.
  3. Selection: a note explaining why two colourways were chosen, and why the others looked wrong in real yarn.
  4. Development: hand-knitted swatches in the chosen colours.
  5. Outcome: photos of the final garment on a model.

The spread makes clear what the student did, and that AI played a small, clearly bounded role.

An invented example for illustration.

Questions to ask before using a tool

  • What do my course rules say?
  • What does the tool’s privacy policy say about what happens to my uploads?
  • What do its terms say about how I can use outputs?
  • What does the company say about its training data?
  • Is there an approved tool my university provides?

Common mistakes

  • Using AI images in place of primary research.
  • Presenting AI images as your own drawings or photographs.
  • Not labelling AI outputs in your sketchbook or portfolio.
  • Designing garments that can’t be made because the AI image didn’t make construction sense.
  • Prompting with a living designer’s name and producing near-copies.
  • Assuming you own the copyright in AI outputs.
  • Uploading confidential internship or client work into a public AI tool.

Checklist

  • I’ve checked my module’s AI rules for this project.
  • My research is based on real sources, places, objects and people.
  • AI images are clearly labelled with tool, date and prompt.
  • My development work shows my own drawing, draping and making.
  • I’ve avoided copying other designers’ work or names in prompts.
  • I’ve read the tool’s terms for using outputs.
  • I haven’t uploaded confidential or client work.
  • I’ve written a short reflection on how I used AI, if required.

Frequently asked questions

Will using AI make my portfolio look worse?

Not necessarily, but a portfolio that relies on AI images without clear process can make it hard for employers to judge your skills. Show your thinking, drawing and making, and label any AI-assisted work honestly.

Can I sell designs I made with AI?

Possibly, depending on the tool’s terms and your local law. But you may not have copyright in purely AI-generated elements, which can make it harder to stop others copying them. Take legal advice for commercial projects.

Which AI tools do fashion designers use?

Tools change quickly. Image generators, AI features in Adobe software and 3D design tools are all used in the industry to different degrees. Your course may provide or recommend specific tools.

How to cite this guide

APA 7Shreya S. (2026, November 17). How to Use AI in the Fashion Design Process (Without Losing Your Voice). Ayerhs Magazine. https://ayerhsmagazine.com/learn/guides/use-ai-in-the-design-process/MLA 9Shreya S. "How to Use AI in the Fashion Design Process (Without Losing Your Voice)." Ayerhs Magazine, 17 Nov. 2026, ayerhsmagazine.com/learn/guides/use-ai-in-the-design-process/.Harvard (check your university’s version)Shreya S. (2026) How to Use AI in the Fashion Design Process (Without Losing Your Voice). Ayerhs Magazine. Available at: https://ayerhsmagazine.com/learn/guides/use-ai-in-the-design-process/ (Accessed: [date you read it]).

Related on Ayerhs: How to use AI in your coursework honestly · How to research a fashion assignment

Sources. Checked 17 November 2026. Formats and policies change, so always follow your own course handbook and tutor first.

  1. US Copyright Office. Copyright and Artificial Intelligence, Part 2: Copyrightability (29 January 2025). copyright.gov
  2. US Court of Appeals for the D.C. Circuit. Thaler v. Perlmutter, No. 23-5233 (18 March 2025). cadc.uscourts.gov
  3. UK Government (DSIT, DCMS and IPO). “Copyright and Artificial Intelligence” consultation (17 December 2024 to 25 February 2025; updated March 2026). gov.uk
  4. Getty Images (US) Inc and others v Stability AI Ltd [2025] EWHC 2863 (Ch), judgment of 4 November 2025. judiciary.uk
  5. Slaughter and May. “Mining the copyright aspects of Getty Images v Stability AI” (2025). slaughterandmay.com
  6. CMS. “Getty Images (US) Inc & Ors v Stability AI Ltd”, AI and copyright case tracker (status at 5 January 2026). cms.law
  7. Copyright, Designs and Patents Act 1988, section 9(3). legislation.gov.uk
  8. University of the Arts London. “AI and assessment”. arts.ac.uk
  9. Russell Group. “Russell Group principles on the use of generative AI tools in education” (July 2023). russellgroup.ac.uk
  10. UNESCO. Guidance for generative AI in education and research (2023). unesco.org
  11. Victoria and Albert Museum. “Search the collections”. collections.vam.ac.uk
  12. The Metropolitan Museum of Art. “The Collection”. metmuseum.org
  13. Adobe. “Adobe Firefly FAQ” (training data). helpx.adobe.com

← All guides · Learn