Beyond AI
How to program with AI without any programming knowledge?

This material has been created in both written and video versions. Watch the video on our Beyond AI channel or continue reading. Thanks!
Watch this material on YouTube:
Hi, I'm Marysia Pożoga. I work at WEBSENSA, a company specializing in AI programming. In my free time, I like to test the capabilities of publicly available AI models and enjoy using them in my daily life.
This time, I was curious about using Chat to help choose clothes for online shopping and selecting outfits for a party. Despite having reliable intuition, sometimes we buy something that has to be returned immediately because it completely clashes with our beauty type or isn't suitable for the occasion.
I noticed that besides professional styling services, there are tons of apps and filters online that try to match colors to our outfits. But honestly? I watched a few shorts, checked out a few apps, and they work... just okay. If I still have to intuitively choose "the flattering one" myself based on colors around my face, it doesn't help me much.

Color analysis for styling involves determining someone's beauty type. According to theory, this is divided into seasons, and each of these seasons can additionally take on attributes related to saturation or contrast.

In a nutshell, to use this for our purposes, we take a photo and ask for a beauty assessment.
First, let's check if the chat correctly recognizes the beauty types of stars who have already been objectively evaluated by styling and color specialists.
I chose 4 photos of famous actresses and asked the chat for an analysis. It handled the beauty of Megan Fox, Julianne Moore, and Reese Witherspoon flawlessly.
Color Analysis of Julienne Moore:

Color Analysis of Megan Fox:

Color Analysis of Reese Witherspoon:

With Jennifer Aniston, its assessment deviated slightly from professional analyses—the chat rated Jennifer as a Light Summer, while colorists usually define her as a Muted Summer.
Color Analysis of Jennifer Aniston:

At first glance—the chat's effectiveness is "quite good."
I sent it my photo asking for an analysis—I should note it was a very clear, well-lit photo without makeup, prepared as professional color analysts suggest.

I must admit that the answers I received are quite consistent and interesting. The chat wasn't as certain as it was with the celebrity photos—after all, it was likely seeing my photo for the first time.

It generated a list of questions and asked for an additional photo in neutral light. It classified me as a Dark/Deep Autumn. According to expert definitions, this type is dark, warm, expressive, and low-contrast. At a glance, that seems about right.

We can assume it correctly assessed the beauty type, so let's see if it can help with shopping—let's give it photos from the internet and ask which of these outfits will suit us. For this trial, I intentionally chose one dress model but in 5 different colors—I'll leave the choice of cut to myself.

I probably would have chosen the light purple dress because it's my favorite color. The chat suggests burgundy or teal-blue, which objectively is probably a better choice for me.
Now let's repeat this with dresses I already have at home—I took photos of them and sent them to the chat. I asked it to pick a dress for an informal party—a friend's birthday.

We see the result—the colorful dress is rated best.Widzimy wynik – tak więc najlepiej oceniana jest ta kolorowa sukienka.

I asked about the others and if they would ever be useful for another occasion. The chat suggested I could keep one, but the other two should probably leave my closet—they don't suit me.

It must be admitted that it handled it very well, but of course, we can consider this subjective and try to confirm it with articles online. Well, the funny thing is that the chat was trained on precisely such articles, so it's hard to expect different results, but it's worth checking.
Let's move to the summary!
If you want to see how to make such an "automated system" for creating descriptions and posting ads, you can see it in Ziemek's episode.
We have many more interesting applications of AI in everyday life in store, so if you want to see something extra, visit our channel—Beyond AI—from time to time.

Yes, ChatGPT is able to perform color analysis based on a photo and determine your beauty type. In many cases, it gives accurate results, though it is not infallible, and it's worth comparing its assessment with specialist opinions.
ChatGPT can effectively match colors and styles to a specific beauty type by analyzing photos of clothes. It can also advise which outfits best suit a particular occasion.
Yes, you can use ChatGPT to help choose clothes when shopping online. It can suggest which colors and styles will be right for you based on photo analysis.
ChatGPT can be a helpful tool in styling, but its advice is based on algorithms rather than the experience and intuition of a professional stylist. It can be great support, but it won't always replace an expert.
AI, like ChatGPT, can be used for many daily tasks, such as creating product descriptions, wardrobe planning, or consulting on various fashion and lifestyle issues.
If you want to learn more about the fascinating world of artificial intelligence and its applications in daily life, visit our YouTube channel – Beyond AI. It is your guide to the dynamic world of AI!

Learn how artificial intelligence can help you unlock YouTube’s full potential and make video publishing easier and more efficient.

Discover how AI can help you pick the perfect present — from personalised recommendations to analysing YouTube reviews and saving time with intelligent suggestions.