Skip to main content
07 / Back to cluster
07 / Disclosure2026-04-136 min read3 sources

Why GOOD CALL Uses AI and Tells You

GOOD CALL uses AI across writing, image creation, planning, and research, and says so in public. The reason is simple: a one-person brand can use modern tools to work faster, but the line has to stay clear. AI can help produce and explain. It does not get to invent facts, fit, or trust.

Definition
AI disclosure

Using AI is not the issue. The issue is whether a brand tells you where the tool ends and where the product facts begin.

GOOD CALL uses AI for writing, planning, research, and imagery. It does not use AI to fabricate reviews, measurements, costs, or operating history.

We use AI. For almost everything. Writing, image creation, planning, research. No in-house agency, no production budget, no team of ten. Just tools used deliberately, and the honesty to say so.

This post explains why. Not as a disclaimer. As a design decision.

01 / Chapter

Efficiency is not a dirty word

Building GOOD CALL as a side project means limited time and limited money. AI closes that gap. A brand with ten employees and a studio budget can spend weeks producing what a solo founder can now produce in days.

Jan note

I can use AI whenever I have time. Modern tools let me work on every topic, everywhere. It still needs careful planning and context management, but I do not have to wait for an agency, a shoot day, or someone to pitch me ideas. I can work fast and keep the investment where I actually want it: in the product.

The alternative is not moral purity. It is slower decisions and more corners cut elsewhere. GOOD CALL would rather disclose the tool and invest the saved time and money in GSM weight, sampling, and factory work.

02 / Chapter

The problem with fashion photography

Most clothing brands show their products on models. That is standard. What usually goes unspoken is that the photos do not just document the garment. They sell aspiration, styling, and a body type.

Even where fashion brands now feature visibly diverse models online, the diversity is usually narrow: in a content analysis of 460 models across 16 brand Instagram feeds, most diverse models displayed only a single diversity trait, and body diversity appeared in only about a third of those posts.SRC 01 That gap matters because it still shapes what buyers think a garment is supposed to do before they read a single measurement.

Jan note

I obviously want to sell you a hoodie. But I want the decision to rest on the product and on what you can actually judge, not on a photoshoot creating desire that the product then has to rescue on arrival.

Audit grid
Signal
What a fashion shoot optimizes
Aspiration, styling, mood, and one specific body type.
What a buyer actually needs
Color, construction, fabric weight, measurements, and trade-offs.
Fit expectation
What a fashion shoot optimizes
How the garment looks on one model under one styling setup.
What a buyer actually needs
Whether the garment is likely to fit your body and your use case.
Decision quality
What a fashion shoot optimizes
Helps someone want the product.
What a buyer actually needs
Helps someone judge the product.
03 / Chapter

What our images are and what they are not

We use two image sources on purpose.

First, we use real photos of physical samples to show material reality: fabric texture, knit structure, stitching, trims, and color behavior in actual light.

Second, we use AI-generated visuals to create additional views and explain the product more clearly across contexts where a full photo set does not exist yet.

The line is simple: real photos document material proof. AI visuals support communication. Neither replaces garment measurements, and neither should be read as a promise of personal fit.

Image boundary
01
They show the product

They can help communicate color, silhouette, design details, and whether the piece reads light or substantial.

02
They do not show your fit

A generated image cannot tell you how the hoodie will sit on your body, your shoulders, or your proportions.

03
They are not a substitute for measurements

GOOD CALL treats fit as a garment-spec question, not as a visual fantasy question.

04 / Chapter

Fit is a measurement problem, not a visual one

Fashion returns are dominated by fit. Around 20% of clothing bought online in the EU is returned, and roughly 70% of those returns come down to poor fit or style.SRC 02 Size charts do not change that arithmetic by much, because M is not a measurement. It is a guess dressed up as a system.

Fit return pressure
70%Roughly seven in ten EU online fashion returns are driven by poor fit or style. Better imagery does not fix that. Better measurement does.

GOOD CALL's answer is simpler than a size-chart theatre performance. Publish the actual garment measurements: chest width, body length, sleeve length, shoulder seam. Then ask someone to compare those numbers to a hoodie they already own and already trust.

What we publish instead
01
Chest width

The dimension that immediately tells you whether the body will feel narrow, regular, or roomy.

02
Body length

The part that tells you where the hem will actually sit, instead of hoping your height matches a vague size recommendation.

03
Sleeve length

Useful because a hoodie can fit in the body and still feel wrong if the sleeve is off.

04
Shoulder and construction context

The measurements only become truly useful when they sit next to fabric weight, knit structure, and fit intent.

The method is not glamorous. It is useful. If the numbers line up with a hoodie you already like, the new one is much more likely to make sense than any stylized product shot ever could.

05 / Chapter

What we do not use AI for

AI helps GOOD CALL produce and communicate. The underlying facts still have to be ours to stand behind.

Hard limits
01
No fake reviews

AI does not get to simulate a customer history the brand does not actually have.

02
No invented fit data

Measurements, size guidance, and garment facts have to come from the product itself, not from a generated guess.

03
No synthetic production track record

If GOOD CALL has not tested something, built something, or audited something, the site should not imply otherwise.

04
No fake cost or transparency numbers

AI can help explain the structure. It does not get to manufacture the numbers.

06 / Chapter

The honest version of transparency

Using AI does not make a brand dishonest. Hiding it does. Most global fashion brands still score poorly on product-level transparency, which is exactly why disclosure at the tool and material level is not a bonus. It is the baseline.SRC 03 GOOD CALL is a one-person operation building a clothing brand in public. AI is part of how that is possible. The public part is not optional.

If that logic makes sense to you, the waitlist is open. And if you want to see the other side of the same transparency promise, the GSM article is already public.

Common questions

Quick answers for the obvious follow-up questions before you move on.

Q01

Do AI images show how the hoodie will fit on my body?

No. The images are there to show color, silhouette, and product details. Fit will be explained through actual garment measurements, not through a model photo pretending to answer a sizing question.

Q02

Do you use AI to invent reviews, measurements, or production data?

No. AI helps with writing, planning, research, and image generation. The numbers, costs, measurements, and operating claims still have to be real and something GOOD CALL can stand behind publicly.

Q03

Why not just do a normal photoshoot?

Because a photoshoot usually sells aspiration before it sells information. GOOD CALL would rather show the product, publish the measurements, and tell you directly where imagery stops being useful.

Source ledger

Every claim in this note ties back to named documents or datasets.

SRC 01
Are diverse models really non-idealized? Investigating body positivity public feed posts of fashion and beauty brands on Instagram
Content analysis of 460 models across 16 brands on Instagram, showing that most visibly diverse models still display only a single diversity trait and that body diversity shows up far less often than racial diversity.
International Journal of Fashion Design, Technology and Education
Open source
SRC 02
Many returned and unsold textiles end up destroyed in Europe
EEA data showing that around 20% of online clothing is returned in the EU and that roughly 70% of those returns are caused by poor fit or style.
European Environment Agency
Open source
SRC 03
Fashion transparency
Reference point for the state of transparency across global fashion brands and why disclosure matters at product level.
Fashion Revolution
Open source
07 / DisclosureWRITTEN BY JAN JERUSALEMAI-ASSISTED / DISCLOSED3 SOURCES