#63 3 Tips How to Deal with AI as a Designer
21:55 Share

Hello friends and welcome back to the future of UX. In this podcast episode we are going to talk about the evolving relationship between designers and AI and I am going to dive into three important tips when it comes to dealing with AI as a designer.

But before we are diving into the topic, I have an amazing announcement because I am hosting a live training next Monday evening. And in this live training, I am sharing my strategy on how to tackle AI. This is a brand new training. And in this training, we will go through the strategy that you need for this year, so for the year 2024.

You're also going to learn about the most common pitfalls for designers when it comes to AI and also how to avoid them. So everything will be very tangible and I will make sure that you get a lot of insights and lessons learned out of this training.

A lot of people already signed up, so I am super, super excited for it. I don't know if I already mentioned that it's for zero euros, so you don't need to pay any money for it. You only need to give me your email address so I can send you all the details and information. And I am so excited for it. It will be a super great webinar. And...

What's also important is if you are busy on Monday evening, don't worry, you can still sign up because I'm going to send out the recording that will be available for 24 hours or 48 hours, something like that. So you can still watch it afterwards and then looking forward to see you in the training.

The podcast episode today is some kind of like the first part of this webinar. So the webinar is the second part, the extended version of this podcast. So in this podcast episode, you are going to see a little or going to hear a little sneak peek of what we will talk about in the webinar. And you can find all the information in the description box. So please check it out.

Yes, but now I think we are ready to dive into the topic of AI and some important tips and updates about using AI as a designer. AI is still here, it will stay and

Suddenly, also people who have been very resistant regarding AI realize that they need to do something now. That this is an important time that really makes the difference on how the future will look like also for their career. AI will change entire industries, not only the design industry, but also HR, management, marketing, basically every industry. Things are changing and to prepare yourself, you really need to understand.

So in this podcast episode, I will talk about three important steps, hopefully some great resources for you. So I would say let's get started with the very first tip. And the very first tip is we think those people or if there are several people who might have a pretty difficult future.

First of all, those who are a little bit overenthusiastic and think AI can solve every problem. It's basically like the new solution to everything. And they are just like overenthusiastic. It describes it very well.

And there are the people who are very skeptical, who think like, ah, this is just a trend, we shouldn't use it. Or the people who are scared, I just got a comment on Instagram the last day or something like that, where someone said, yeah, we should like as a community block AI and not use it.

This is not going to work. This is not going to work. Because if we would even say like we are blocking AI out of our world, then other industries would take over. And the developers, the business sector would design suddenly products and experiences. And we don't want to do that. We fought so hard that we as designers get a seat at the table. So we need to be part of that movement now and understand. And only if we really understand, we can make good decisions.

And one thing that I definitely want to highlight is AI is this, you know, like this huge buzzword that you're hearing everywhere. But I think you could imagine AI a little bit like a coworker, I always say that, or like a little assistant. It's not your calculator, right? Where you just like put something in and then get something out.

But it's really much more an assistant than anything else. And the thing with a new assistant or with a new co-worker, you need to train them first, right? And you need to build trust. And especially if this is someone who's like maybe a beginner, like a junior assistant very early in their career, you need to double check certain things. And also these assistants need to go to gradual training and adaptation.

AI generally is very good at executing certain tasks. When we look back at the history of humankind, of the work environment, when we needed a lot of factory workers, the whole topic of execution has been super important. Getting a task and then getting it done. Not to think about other solutions left and right, but just execution. Executing certain tasks.

And now things have changed, right? Because of all the automation and the internet and the focus shifted much more on re-ideation, coming up with problems and all these human skills where you really think. So when you think about the different steps from like having a certain problem to getting to a solution, there are a lot of steps in between, right?

First of all, you define the problem. You really think what is the problem. Then you put certain resources in it. You produce an outcome and you produce an output. And then this produces an outcome. And then you have the solution, the result that comes out of this. And I think what's super important is to understand where AI plays a part of these new AI workflows. And this is for sure not...

presenting you the problem. But this is something that a human needs to do, coming up with the problem, ideating. But with everything that gets into the direction of execution, AI is amazing and can help you so much.

So what you need to do right after you came up with a problem, you defined the problem statement, you have the briefing, you maybe put your first thoughts in it. You also use AI to organize these thoughts, to prompt it with background information, with what are the next steps and then let AI produce something, an output basically, right? Like a text, a prioritization board, anything like that.

and you use these outputs to iterate. Of course, you know, like with also a junior assistant, you know that this is not super perfect. You need to iterate on it and change certain things and certain topics and then go through the iteration again. And I think this is probably one of the most important things at the moment, really understanding how roles are changing also for designers.

I'm also seeing that more and more when I'm talking to my clients. As you know, I'm working as a freelance designer. So I'm working with many different companies and clients from all around the world. And of course, also my clients get excited about AI and want to use certain tools. For example, a few days ago, a client called me or actually sent me an email and sent me a link with a Galileo AI link. So he prepared a prototype.

like just on its own using a single prompt and entering it into the Galileo AI tool, copying that to Figma and then having everything in Figma. And he was so proud of himself, like super excited, sent it to me and said like, oh, wow, Patricia, look what I did. It's like so fascinating. Maybe we don't really need a proper UI designer anymore. That's totally fine because we were actually looking for a UI designer who could do the UI work.

I was like, okay, interesting. Amazing. Thank you so much for trying these things out and keeping me in the loop. I think it's a very good first step. Maybe we can use a

like a few ideas here and there and then iterate on that, right? So also a lot of communication between the client, AI, how to use it. So there's a lot of back and forth and I see also designers changing more into the role of being the navigator. So what is important when it comes to rethinking? I think rethink, especially when it comes to roles that are changing.

I think it's super important to really stay updated. Updated on the current tools and technologies. Even if you are not a super techie or futuristic person, AI will be a part of everything. If you are a researcher, if you are a UI designer, if you are a brand designer, an illustrator or a UX designer,

AI will become a huge part of all these areas. And the better you know what's possible, the better you can use it because also expectations are changing. And we'll talk a little bit more about that later on in the podcast. So the biggest takeaway for the first part is really stay updated with the current tools. Designers especially or creatives are the perfect target group when it comes to AI tools because they can really use it and leverage it.

This brings us to topic number two, which is really leveraging AI for design. And this is what I am currently seeing and I think it's pretty fascinating. AI tools are especially empowering also solo designers, small teams or if you're working on your own. Suddenly, also as a solo designer, you will be able to

get huge amounts of work done because you have tools who do all the automated tasks for you, all the execution basically. So there's one thing that's super important for you to understand when it comes to leveraging Generative AI as a designer. First of all,

understand what are the parts that can be automated, things that I'm doing on a day-to-day basis, things that just can be automated or what I can use AI for and what are the things that I should probably do on my own. And I'm also going to give you an example now.

Let's start with Chachibiti, a tool that we all know. Chachibiti is very good when it comes to data understanding. For example, imagine you have a research project right in front of you or a design problem that you want to solve how to integrate AI. AI can be very helpful when it comes to

preparing the research session, so really coming up with research questions that are aligned with the long-term goal and the vision and the the question that you actually want to have answered. Important there is of course the background information, so to prime the GPT, give them all the information that can be a role, the set of context, constraints, expectations, maybe details, maybe examples.

and then ask to prepare the research question. And they might not be perfect, but you can use that as a starting point and then iterate on that. So what AI is not so good at, and I have seen that several times, but I think it's still interesting to see, are creating anything where you need that kind of empathy.

For example, creating user personas, for example, with ChatGPT or other AI tools. You can do that and you can try it out and then make your own assumptions or see if it's working for you. But most of the time, from my experience, those personas are not really what you need.

Because they are a little bit made up and what I'm seeing with AI is that also with personas, you need to fill the gaps a little bit, right? You need a little bit of imagination to come up with these personas and AI gets sometimes a little bit too creative with these. So what you can always understand and use, I think, as a guideline is if you put a question into TedGPT, is it something that's focused on execution?

If yes, okay great, then this is perfect. Is this something that's where you need empathy for? Understanding a certain problem space, talking to users, these kind of things. AI is probably not the perfect fit for it. And is there some kind of background information that you can't put into words that are that's constantly changing for example? If that's the case, AI is also not the best tool for it.

So as a little takeaway: Generally AI is amazing at finding patterns. So when you have these research projects, you have all these interviews, you put them in and you ask to find patterns, that's amazing, super helpful. Now let's move to topic number three, which moves a little bit into the, I would say, like future directions and next steps.

And as I mentioned, there is a huge shift coming in professional roles from executing tasks to validating AI outputs and setting constraints. Because what definitely is a huge challenge to validate the output, as you have all probably recognized, you put something into TechGPT or another tool, you get an output, you can produce like a million texts, a million images, and then you need to validate them.

And in the beginning, you also need to set the right constraints, like how should this text be? What does it doesn't should be? So these kind of things. So I'm seeing a growing demand for professionals who can effectively leverage AI, who can use them, who become more productive and use it much more strategically.

Clients expectations are already changing and I'm seeing that I'm talking to a lot of different people in the industry. And of course, if you have a professional copywriter, someone who writes your blog posts or social media, when you gave them a task,

a year ago or two years ago and you said okay this is around like six hours or so now you would expect them to do it in two hours and you would automatically expect them to use chgpt if you agreed on that so it's much faster and you also expect higher quality like no typos basically right because they can always check it with ai tools so my tip here for you is to really explore things at the moment things are pretty quick outdated and also the tools are

changing all the time so if you are a total beginner and you feel like oh there's so much going on i feel really overwhelmed with all the ai hype i maybe can't even use that in my own company it's forbidden we can't use that my tip for you is this

Think about a problem that you are having currently. That can be a design problem. That can be a certain topic that you are excited about. And create your own mini project around these problems. Start with AI tools. The easiest and what I always recommend is to start with ChatGPT or Google Gemini or Cloud or Microsoft Bing.

as a text generation tool and then choose one visual generation tool like mid journey or adobe firefly for example and get started with those. don't feel overwhelmed but start steadily by using these tools. and here comes the biggest tip: try

Try to integrate them in your current workflows. That might be: use Midjourney in combination with Magnific AI or Photoshop. This is what I'm always doing. So when I am using these tools, it's not just that I am creating images of Midjourney, but I'm using them in certain different workflows. A lot with After Effects, for example. I'm going to share more about that on Instagram in the next couple of weeks.

So this is an amazing combination, right? Or with Photoshop that you can edit them further. So don't try to solve every problem, everything that you want to achieve with one single tool. This is not going to work. You really need to think in workflows. What other tools can be combined with them to get the output or the results that you are desiring? Okay, so my little tip here or advice is to

really try things out, explore a lot and think about the different workflows where I fit in.

So this has been a quick intro or a little sneak peek of the webinar that I'm going to host next Monday evening. I'm super excited about it. I'm going to share much more examples, a lot of free sneak peeks into my own workflow. So it will be super, super interesting. And I highly recommend you to check it out and sign up, join us. The last webinar in January has been amazing. So I would definitely love to see you in the webinar and see

yeah talk about AI and get ready for the future so you can find all the information in the description box sign up it's free and then the only thing that I can say or that is left to say thank you so much for listening and hear you in the future