Jennifer Mackey-Mary: Welcome back to another installment of “Are the machines taking over the style world”. In this episode, you’ll discover if ChatGPT can replace your favorite stylist. I know you’re dying to know, so let’s get started. 

Welcome to the Everyday Style School, the show that teaches you everything your mom never did about getting dressed.  I'm your host, Jennifer Mackey Mary. After 25 years of dressing women with real bodies, real budgets, and real lives, I know great style isn't about following one-size-fits-all advice - it's about learning what works for you.

Hello Gorgeous! 

Here we are, part 2 of Can AI make style easier? In the first episode, I talked about those AI style apps you’re constantly seeing on your social media feeds. It’s funny, I was talking to a group of women at an event last week, and someone asked what I was working on, and I told her about these episodes, and she didn’t really know what I was talking about, so I said–you know those ads you see on social media for the 10 piece amazon capsule wardrobe, and she was like “Those are AI? I had no idea”, and if we’re being honest, I bet a lot of us see things and don’t realize they’re not real, or really created by humans anyway. I will fully admit to sending my husband cute baby penguin videos, and then having to send a follow-up text that says “never mind, that’s not real”. Point is, AI is creeping in to all corners of our lives, and it’s getting harder and harder to tell the difference, so It’s important to be aware, and educated, and that’s what these episodes are all about.

If you haven’t listened to part one yet, I really recommend going to check that out. The truth is, you’re probably more likely to use the apps for style help, which is what I talked about in that episode, so don’t skip it. 

In this episode, we’re talking about what we usually think of when we say AI–Large Language Models like ChatGPT, Claude, Gemini–that kind of thing. I think these are important to talk about in the style arena because a lot of people are going straight to these tools to get information, instead of using Google or another search engine. They work differently–ai gives you the answer, search engines show you where you can get the answer.

Like most everything in life, there are pros and cons to both. It’s so much easier and more efficient to just get the answer, right–as opposed to going to each site, realizing that’s not quite what you’re looking for, going to another one. It’s definitely more convenient to just get what you’re looking for. But on the other hand, a single source of truth is only as good as that single source of truth, and that’s what we’re going to get into today. 

How good are they? Can this type of AI make style easier or better? Let’s find out.

Remember, to form my opinions on ai style apps, i relied on reviews and hours of youtube videos. For this one though, I went to the source, and actually did the thing. When I started, I posed the same 2 questions to 3 models–Claude, Gemini and ChatGPT. 

I asked “What styles are best for apple shaped women” and then told them to create a 20 piece Capsule Wardrobe for an apple shape. I know the prompts were simple and somewhat vague, which was on purpose. 

A lot of the responses about what styles were best for an apple were kind of odd–everything was about elongating the torso, which as an apple with a long torso, I was confused by. That’s a vertical body shape thing, not necessarily a horizontal body shape thing, and it was just strange how much of the reasoning behind the advice was about torso length. But otherwise, it was pretty standard, here’s what an apple should wear advice. There were some other issues with the style advice, which we’ll get to later.

The Capsule results had some overlap because the advice on what styles were best had overlap, but gemini and chatgpt gave me 20 clothing pieces, while claude gave me 16 clothing pieces and 4 accessories, which made for a much more limited capsule. Gemini tried to give me a wardrobe for all seasons, which again, made it less functional, and all 3 seemed to believe a capsule wardrobe simply meant all of the pieces were in the same limited, neutral palette. Which, to be fair, when you have a capsule that small, you kind of have to have a limited neutral palette. But there’s more to capsules than color. Do the silhouettes of the tops work with all of the pant silhouettes–that kind of thing. That was clearly not a consideration. 

In the end, all three were pretty similar in terms of the quality of the responses- there wasn’t a real clear winner, but I thought ChatGpt was slightly better, so I chose to use that for the rest of my research. 

For the next part of my little project, I went to our community, pulled questions and photos from actual members (don’t worry, I cropped out faces and identifying info), and hit enter to see how my answers–that I feel very confident in-would compare. 

In the last episode, I went over some of the common app features, and at the end gave my overall thoughts, but today, I’m just going to turn to the last page in the book, and give you my final thoughts. Then I’ll tell you why I think that. Here goes–overall, this is bad. It’s not good. Please do not rely on tools like ChatGPT to reliably make your style easier or better. I’ve got five reasons why. Here goes.

#1–The goal of AI is to make you happy, not necessarily to be right. This is something I heard on one of my favorite podcasts over the summer–and it really changed the way I look at these tools. Have you ever noticed how it agrees with EVERYTHING you say, and validates you no matter what? That’s what this is all about–making you feel good. And what happens when we feel good? We do more of it. Like social media It’s designed to get you addicted to using it, often at the expense of truth and good information. To me this is really counter intuitive, because you know what makes ME happy? Correct information. That makes me happy. I guess I’m weird. Anyway, I’ll tell you what that looks like in the style arena, but first, a

Quick sidebar-I want to share HOW I use chatgpt in my business, because it matters here. I don’t use AI to write my content, for a lot of reasons. But I will use it for things like project management-if I’m working on a course, I’ll give it a brain dump and let it create sections that make sense to not just me, and it will suggest things I might be missing. I’ll also use it for research and brainstorming–like if I’m doing an episode on I don’t know–10 ways to improve your style for free, I’ll say. Here are my ideas, give me 50 more ways to improve your style for free. I always know what I want to say, but in case you haven’t noticed, I’m human, and I don’t want to forget something that’s important and obvious, just because my brain didn’t come up with it at that moment. I always ask for a ridiculous amount of responses, and Normally they’re really stupid, but every once in a while, it will be like, How did I miss that, or it sparks something else I want to say. Usually not useful, but now and then, it is. Anyway, I tell you this because over time, ChatGPT has learned my business, and when I started my little research project, I could tell it was using my ideas, my voice, my wording. But I know that’s not the experience everyone is going to have, so for this, I opened an incognito window, and the results were VASTLY different. What you get, is NOT the same as what I get.

When it comes to making you happy, instead of defaulting to correct info, here’s how that played out. I gave it pictures of outfits, and the question the member asked, like is this flattering, or is this current. In my usual ChatGpt, the responses were pretty decent, maybe 75% of the way there. In the incognito window, which mirrors the experience of someone who doesn’t feed it style advice on a regular basis, almost every response to “Is this flattering? Or “is this current”” was yes. At the end, it would say, “if you’d like me to suggest one small tweak to make it more flattering, I can do that”, but it most certainly didn’t lead with that. And even when I said “yes, give me the tweak”, it softened it quite a bit. It was definitely on the side of making the user feel good. So the next time you use ChatGPT, and it tells you you’re the smartest, prettiest, wittiest woman in the land…take it with a grain of salt. I mean, I’m sure you are, but… you get what I’m saying. If you’re looking for brutal honesty about your outfit, this isn’t the place to get it.

Issue #2–These platforms make stuff up. There are lots of ways AI gets things wrong, but this is by far my favorite, and by favorite, I mean the most hilariously frustrating. If you’re not aware of AI Hallucinations, here’s what they are–here’s the definition straight from google: AI hallucination is a phenomenon where artificial intelligence, particularly large language models (LLMs), generates confident, plausible-sounding, but entirely false, nonsensical, or ungrounded information.

We’re not talking about referencing slightly outdated information, or otherwise misunderstanding the prompt. This is straight up fabrication. 

A while ago, I was working on an episode, and I wanted to include some research studies, so I asked ChatGPT to find me peer-reviewed scholarly articles on the topic, and pull relevant quotes. Way faster than doing it myself, right? I think the episode was something about clothes and emotions. In seconds, I had a ton of studies, lots of quotes, including one that said something like -”Our emotions are like clothes hanging in our closets. By choosing the correct ones, we can become the most beautiful versions of our inner selves and change the world for the better”, and it was attributed to someone with PHD after her name, we’ll call her Sarah Smith. Now, if you’ve ever read a research article, you know that doesn’t sound like something you’d find, right? There’s no flowery, poetic language. They’re dry and boring. So I asked–can you show the full article from Dr Sarah Smith, and the response was “No, I made it up. Sarah Smith doesn’t exist”. Um, no…no. We don’t do that. 

Now, in all fairness, I didn’t come across any hallucinations while doing this project, but for me, its enough to know there’s a possibility. You could ask ChatGPT what’s trending this season, and it tells you that because of the Artemis mission, there’s a nod toward space, with metallic jumpsuits and moon boots being hot this season. Which is definitely not a thing

But, this doesn’t mean all the answers are totally made up or even wrong at all, which ironically is my third issue with them.

If the information you get from things like ChatGPT was right all the time, we wouldn’t have a problem. You could rely on it. And if the information was wrong all the time, you wouldn’t have an issue, you could just ignore it. The problem is that it is right some of the time, but not all of the time, and unless you have the skills to know which is which, you can’t trust any of it. Imagine I invited you over for a party, and had a big table of food laid out and said, 70% of these dishes are great. The other 30% will make you sick. And you have to guess which is which. You wouldn’t trust any of it, would you? You’re driving through McDonalds on the way home, because you didn’t eat a thing/

Same thing here–If you don’t have enough foundational knowledge to tell which advice is good, and which is bad, you can’t trust any of it. Last episode, I briefly touched on an app that promises body shape and color analysis, and a big criticism was accuracy. Some of the social media comments said “it was spot on”, but enough said “this was way off” to plant that seed of doubt. If you don’t know what your body shape is, how do you know if you got good info?

A while ago, I got an email from someone- she’s a social media expert, and this had nothing to do with her business, But it said I found the greatest life hack you have to try. And the hack was using ChatGPT for color analysis. She said Just take a picture, upload it and ask chatgpt to tell you your season. Easy, peasy, free. Later that day, I was meeting with our Capsule Creative director, and we were talking about it–she was said, yeah, I did that, it was great. So I tried it. Wrong. How do I know it was wrong? A couple of ways–I have enough color knowledge to question the results…but also, I’ve been analysed by every single color system out there. Whatever their version of the lightest spring is, that’s what I am. Call it by any name, I’m light, bright and warm. The colors on those color cards are my best colors. But if I didn’t already have that knowledge, and just took ChatGPT at it’s word, I’d be walking around in some dusty blues that make me look paler, older, and more tired than I really am. No thanks.

What I realized though, is that both the business owner, and our creative director have very easy colors to determine. There’s no question. Even I could do their color analysis, which is saying something. I can tell you HOW to find your colors, but I lack that eye. My coloring is a little trickier. My eyes are warm, my skin is neutral warm, but my hair is naturally very cool. So I decided to give ChatGPT another crack at it, with another difficult to type test subject–my youngest daughter. She has dark dark brown eyes, very fair skin that’s clearly cool, and medium brown hair with golden highlights. A couple of years ago, She wanted her colors done, so we did it, and even the professional said yeah, she’s tough. But eventually we figured it out. ChatGPT was so far off the mark, it was ridiculous. It recommended her worst colors, and what I realized is that the harder you are to categorize, the greater the chance the answer will be incorrect. And here’s why that matters: the easier you are to figure out, the less help you need to figure it out. Its those people who are hard to type who turn to professionals, or in this case the gnome who lives in their computer, for help. And if you don’t know how to determine if those results are right or wrong, what are you going to do? If you can’t trust it, what’s the point of doing it?

So back to my research project–the first photo and question I put into ChatGPT for this project was a member who wanted feedback on flattery and currentness of an outfit. The response was perfect. It said everything I would have said…actually, this is the one that made me go incognito it was that much like me. But even the incognito version got it right. I had a moment of “that’s it, I’m being replaced”. I discourage easily. Anyway, I did another one–another member asking about flattery and currentness. In this one, there was ONE thing that was throwing everything off. One obvious issue. Yeah, some minor tweaks, or “you could do this” but to fix the outfit, I would have fixed one thing. My chat and incognito chat both missed it. Gave a list of other things to fix, but not the one thing that would have made a difference. Again, if you're someone who doesn’t know what’s wrong with the outfit just by looking at it, which is the reason you’re asking in the first place, you can’t know if the results are right or not. And if you have no idea how often it’s wrong, you might be tempted to trust it. 

But, getting bad advice isn’t always AI’s fault–sometimes its yours, which is issue #4

The advice you get is only as good as the prompt you give, and most people don’t know how to give the right prompt to get what they need. 

Going back to the try this life hack email I just told you about, the prompt was “Evaluate the colors of my eyes, hair, and skin and tell me which of the 12 color seasons I fall into”. That was it. 

Last fall, I created a bonus course for the Style Circle members who joined for a year during our holiday sale that was all about finding your best colors. Again, I can’t tell you what your best colors are, but I can teach you how to find them. I’m the literal definition of those who can, do and those who can’t, teach. I’m fine with that. Anyway. In this course, I give I think 4 ways to find your color season, and encourage you to do ALL of them, because chances are, one way will click, or there will be enough similarity in results that you can get to the right answer. It’s the best way to DIY something like this. One of these ways is using a platform like chatgpt, so I included a prompt. My prompt is six pages long. Yes, you heard that right–6 pages. There’s the initial 3 page prompt, which includes a questionnaire for ChatGPT to ask you so it’s not just relying on one or two photos and then it has a detailed format to deliver results so you can check for accuracy step by step. It’s then followed by 3 more pages of follow up prompts to verify results, explain the why behind the answers, compare secondary seasons, and more. Which do you think is likely to get better results–one sentence, or 6 pages of instructions? 

But most people don’t give instructions that detailed. And why would you? Who has time for that? I read somewhere that the perfect prompt takes 8 hours to create, and I definitely spent that. But the average user isn’t going to do that for themselves. At that point, Just hire someone to tell you your colors–what’s your time worth. And to be honest, most people don’t know HOW to create a prompt like that because If you knew everything you needed to ask, and knew during testing where it’s going wrong, you’re probably far enough along that you don’t need someone to give you the answer. 

I’m not saying that every question needs a crazy long perfect prompt, but most of the time you can’t give it one sentence and expect good results, which is what we’ve been trained to do through search engines. You can’t use them the same way.

And it’s not always about the prompt either–all of these style platforms depend on your photos–whether we’re talking about an outfit idea from an app, or a body shape analysis on ChatGPT. I saw one review video where the woman did notice that the items the app suggested didn’t really go together, and she was wearing the outfit in the video, and she showed the screengrab of the app created outfit on her phone–they looked totally different because of lighting. In the app, they went together, in real life, not so much, and not the apps fault.

The same thing happened during my ChatGPT color analysis. When I was putting together and testing my 6 page prompt, I tested the photos I had used before–you know, when I got the wrong results. I was trying to figure out if a better prompt got me to the correct answer. It didn’t, But this time I said, I’ve been typed as a light spring multiple times, why are you giving me a cool summer result. It analysed the photos again and asked me “were these photos taken in winter?”, and sure enough, they were. I followed all the instructions–no makeup, facing a window, on a sunny day–but the winter sky added a gray coolness that it picked up on. Not really my fault, I hadn’t even considered that. Not chatGPT’s fault–it just used what I gave. 

I also found this with the photos from the community I tested with. One thing my style circle members hear me say all the time is that I need good light in the photos they submit for style coaching, because its hard to give feedback to someone in an all black outfit in a dark room. I can’t see what I’m giving advice on. I gave it one of those photos just the way it was submitted, and then compared it to the advice I would give–but for me, I lightened the photo in canva enough that I could see details–and ChatGPT’s advice was way off. It was valid advice for the dark photo, but not the real situation. 

My last issue is that they often give accurate, but dated, advice. When I asked my original question about best styles for apples–all of them gave apple advice straight out of 2013, suggesting leggings and tunics. To understand why this happens, you have to understand how Large Language Models get their answers. Unlike a search engine, where you ask a question, and gives you a list of resources, ai looks for patterns in what it’s been given in the training phase of the model, and what users have fed it since then. I told you I don’t use AI to write content, and this is one reason why. I’m not giving it my ideas. I have turned off the setting in ChatGPT that allows it to use my inputs for training. I’m not helping it give better style answers to the world. Sorry. Those are my ideas. Go do your own work.

I had a very meta conversation about this with chatGPT, understanding exactly where the answers come from, and the last thing I asked was “how do you stay up to date”--do you scour the internet looking for current information? Here’s the answer–still in incognito mode…

“That’s a great question, and you’re wise to zero in on an important distinction”--thats making me feel good—  Then it said, usually no. By default, I don’t browse the internet in real time when answering. I generate responses based on what I learned during training and in general patterns. For something like apple shaped body styling, the core ideas haven’t changed much in years, so older info is still relevant. However, if you ask something that is time based like trends for the current season, I can look up current info and incorporate it. 

So it’s using all the blogs written a decade ago as the basis for it’s answers. That’s why the capsule wardrobes they created for me look like I got in my time machine and went back to my closet when I started my business in 2012. All of those things were in here. A lot of the PRINCIPLES of dressing an apple shaped body are the same, but current STYLES have changed drastically, and we’ve had to change the styles we use to keep those principles the same–if that makes sense. So its not really the premise that’s off, It’s the execution, and causes the end result to miss the mark–and isn’t that why people use platforms like this in the first place–to get that quick done for you answer? If the information you get isn’t useable, again, what’s the point?

So, here’s quick recap of the cons of using ai models for style help–they’re more focused on making you happy than they are on giving good answers, the accuracy can’t be trusted–and that can be because they just straight up fabricate answers, they don’t get good inputs, or because they reference accurate, but dated information. All of that leads to variable accuracy, which is the overall issue. When you don’t know if the information is good or not, you simply can’t trust it. Now, I said I was going to give pros and cons to both–and yall I am trying to find the pros here, and I guess the pro is that it’s easy, free, and accessible, and sometimes very good–but again, if you don’t know if you’ve gotten good advice, what’s the point? Im sorry, Im really trying to be fair and balanced, but I keep coming back to this idea that the people who rely on tools like this the most are those with the least amount of knowledge to tell which advice is valid and which isn’t. To me, that’s just a recipe for more style confusion.  

Before we wrap up, I want to share one final thing about using AI to improve your style, and it’s something I touched on in the last episode, when I said that Giving you outfit ideas each day makes it easier to get dressed in the moment, but it can’t replace the skills you learn from picking a top out of your closet and challenging yourself to make 5 outfits with it. Remember that? That right there is the biggest problem I have with using AI for style help. It doesn’t teach you anything long term. But, really, it’s worse than that. 

The more you rely on AI for something, the worse you get at that thing. It’s not just that it doesn’t make you better, it actually makes you worse. Here’s how: First, through automation bias. Studies have found that humans are more likely to accept incorrect answers when they come from an automated system, than when they come from a human. Humans we question, machines we don’t. Doesn’t sound problematic at all. Over time, you stop using your own judgement, and the answer to so many style questions is, “What do YOU think of it? How do YOU feel in it”. Your own judgement is vital when it comes to style, don’t give it away.

Second, through lowered creativity. A study at MIT Sloan school of Management found that AI helps people generate ideas faster, but the ideas tend to be more similar to each other, and less original overall. And the more people used AI, the less able they were to come up with new ideas on their own. I’d argue creativity is pretty important when it comes to style. Again, don’t give yours away.

Finally, Through cognitive offloading, also known as the GPS effect. Ever notice that you used to be able to know how to get places, but after years of relying on your GPS, you don’t know how to find anything? You’re not getting old, you’re experiencing cognitive offloading, which is that when people rely on tools to do thinking for them, they practice those skills less, and their ability in that area weakens over time. 

To put this plainly, the effects of relying on AI are that you trust it more, you think less creatively, and over time you get worse at doing it yourself. This whole show could have been that one sentence. 

Let’s be real for a second, I don’t think any of you are currently addicted to using ChatGPT or a style app to get dressed. You probably use it occasionally, which isn’t a big deal, but like with all good drugs, that’s where it starts. And even if you’re not feeding an addiction, you still have the issue with reliability, which in the end, gives AI a big thumbs down for me. 

Your homework for this episode is to do nothing. And by that, I mean if you’ve been listening to this episode, thinking, maybe I rely on AI too much for anything–doesn’t have to be style relate, stop. Break up with it. Challenge yourself to do the work on your own. Build that muscle back up. 

OK, AI lovers, grab your pitchforks and light the torches. I’m ready for you. I know that this was a very simplistic look at AI and all the things it can do. There are absolutely great uses for AI, and amazing innovation happening. But I’m a wardrobe stylist dedicated to helping women build lifelong style skills, and today, I don’t think AI is the way to get there. 

That’s it for this episode of the Everyday Style School. Thank you for spending time with me today.

If you’re ready for easier, better style, come take my free workshop Style Made Simple, where you’ll discover how to stop relying on tools that don’t teach you anything and start relying on your own knowledge and opinions so you can make style easy for real— for life. 

You can sign up at freestyleclass.com or through the link in the shownotes.

I’ll see you next time–and until then, stay stylish!