I’m going to be blunt. Most of the AI-generated design I see is garbage. It’s a sea of generic, plastic-looking visuals that all scream “I was made by a robot.” After two solid years of going deep down the rabbit hole of AI-driven design, I can tell you that the tools themselves are not the problem. The problem is how people are using them.
I’ve built and sold two companies, RemoteTeam and MovieLaLa. I’ve invested in over 200 startups, including some of the biggest names in AI like Anthropic, OpenAI, and Scale AI. I’ve seen what it takes to build successful products, and I can tell you that design is not something you can just automate away. But you can use AI to supercharge your design process, if you know what you’re doing.
I’ve spent thousands of hours and probably a small fortune on subscriptions to Midjourney, DALL-E, Stable Diffusion, and a dozen other tools. I’ve run countless experiments, trying to create everything from logos for my portfolio companies to marketing materials for my book, "Becoming Top 1%". I’ve made a lot of mistakes along the way. But I’ve also had some big wins.
This isn’t another guide on how to write the perfect prompt. This is about the real, hard-won lessons I’ve learned about what it actually takes to create professional-grade, AI-driven design. These are the 7 things I wish I knew when I started.
1. Your Taste is Your Most Valuable Asset
This is the single most important lesson. You can be a master prompter, a wizard with every setting and parameter. But if you have bad taste, you will create bad art. It’s that simple. The AI is a tool, a very powerful one, but it’s not a replacement for your own creative vision.
I learned this the hard way. When I first started, I was obsessed with technical perfection. I would spend hours trying to get the lighting just right, or the composition perfectly balanced. But the images I was creating were lifeless. They were technically impressive, but they had no soul.
It wasn’t until I started to inject my own personality and taste into the process that things started to click. I stopped trying to create what I thought the AI wanted to create, and I started trying to create what I wanted to create. I started using my own experiences, my own memories, my own emotions as inspiration. And that’s when the magic started to happen.
One of my angel investments, a small gaming studio, was struggling to define the visual identity for their new title. They had a team of talented artists, but they were stuck in a rut. I suggested they try using Midjourney, not to generate final assets, but to explore different art styles. I told them to forget about the technical details and just focus on creating images that they found beautiful and inspiring. The results were incredible. In a single afternoon, they generated more creative ideas than they had in the previous month. The AI didn’t replace the artists; it unlocked their creativity.
2. “Garbage In, Garbage Out” is a Myth
Everyone says that the key to good AI art is a good prompt. "Garbage in, garbage out." That’s only half true. The real key is iteration. It’s about starting with a simple idea and then refining it over and over again until you get something you’re happy with.
My best creations rarely come from a single, perfect prompt. They come from a process of trial and error. I’ll start with a very simple prompt, maybe just a few words. Then I’ll look at the results and see what I like and what I don’t like. I’ll take the parts I like and add them to the prompt, and I’ll take the parts I don’t like and I’ll try to get rid of them. I’ll do this over and over again, sometimes dozens of times, until I have an image that I’m proud of.
For example, I was recently working on a logo for a new startup. I started with the prompt “a minimalist logo for a tech company.” The results were… predictable. A bunch of generic-looking icons and wordmarks. So I started to iterate. I added more details to the prompt. I tried different styles. I experimented with different color palettes. After about 50 iterations, I had a logo that was unique, memorable, and perfectly captured the essence of the brand.
Don’t be afraid to start with a “garbage” prompt. The important thing is to start. The magic is in the iteration.
3. Master One Tool, Then Diversify
The AI art world is a confusing place. There are dozens of different tools to choose from, each with its own strengths and weaknesses. It’s easy to get overwhelmed.
My advice is to pick one tool and master it. For me, that tool was Midjourney. I spent months learning the ins and outs of the platform. I read every tutorial, watched every video, and joined every Discord community I could find. I learned how to write effective prompts, how to use the different parameters, and how to get the most out of the platform.
Only after I felt like I had truly mastered Midjourney did I start to explore other tools. I started playing around with DALL-E 3 for its incredible photorealism and its ability to understand natural language. I experimented with Stable Diffusion for its open-source flexibility and the power of its community-built models.
Today, I use all three tools on a regular basis. I use Midjourney for its artistic and stylistic range. I use DALL-E for its photorealism and its ability to create images that are almost indistinguishable from real photographs. And I use Stable Diffusion for its power and flexibility.
But I wouldn’t have been able to do any of that if I hadn’t first taken the time to master one tool. So pick one, and go deep.
4. The Ethics Are Your Responsibility
AI art is a new and powerful technology, and with great power comes great responsibility. There are a lot of ethical questions surrounding AI art. Who owns the copyright to an AI-generated image? Is it ethical to use an AI to create art in the style of another artist? What happens when AI is used to create deepfakes and other forms of misinformation?
These are not easy questions, and there are no easy answers. But as a creator, it’s your responsibility to think about these issues and to make sure that you are using AI in an ethical and responsible way.
For me, that means being transparent about my use of AI. It means respecting the intellectual property of other artists. And it means being mindful of the potential for my work to be used in harmful ways.
I’ve had to make some tough calls. I was once asked to create a series of images for a political campaign. I turned down the project because I was not comfortable with the idea of my work being used to influence an election. It was a difficult decision, but I believe it was the right one.
5. The Real Work Starts After the Generation
Generating an image with AI is the easy part. The real work starts after the generation. That’s when you take the raw output from the AI and you turn it into something truly special.
I almost never use an AI-generated image as-is. I always bring it into Photoshop or another image editor to do some post-processing. I’ll adjust the colors, tweak the composition, and add my own personal touches. Sometimes I’ll even combine multiple AI-generated images to create something completely new.
For the cover of my book, "Becoming Top 1%", I went through hundreds of AI-generated images. I finally found one that I liked, but it wasn’t perfect. The colors were a little off, and the composition was a bit weak. So I brought it into Photoshop and I went to work. I spent hours adjusting the colors, cropping the image, and adding text. The final result was a cover that I was incredibly proud of, and it was a perfect example of how you can use AI as a starting point to create something truly unique.
6. Constraints Are Your Best Friend
One of the most counter-intuitive lessons I’ve learned is that constraints can actually boost your creativity. When you have unlimited options, it’s easy to get paralyzed by choice. But when you have to work within a set of constraints, it forces you to be more creative and resourceful.
I learned this lesson when I was working on a project for a non-profit. They had a very limited budget, so I had to be creative. I couldn’t afford to hire a photographer, so I decided to use AI to generate all of the images for their website. But I didn’t just generate random images. I created a set of strict constraints for myself. I decided that all of the images had to be in black and white. They all had to have a very specific aspect ratio. And they all had to feature a single, isolated subject.
The constraints forced me to be creative. I had to think outside the box. And the results were some of the most powerful and moving images I’ve ever created.
So don’t be afraid of constraints. Embrace them. They might just be the key to unlocking your best work.
7. The Human Designer is Not Going Away
There’s a lot of fear and hype out there about AI taking over the world. People are worried that AI is going to replace artists, designers, and other creative professionals. I’m not one of them.
I believe that AI is a tool, and like any tool, it can be used for good or for bad. It can be used to create generic, soulless art. Or it can be used to create art that is beautiful, meaningful, and deeply human.
The future of design is not about AI replacing humans. It’s about humans and AI working together. It’s about designers using AI to supercharge their creativity, to explore new ideas, and to create things that were never before possible.
I’ve seen this firsthand in my own work. I’m not a designer by trade. But with the help of AI, I’ve been able to create designs that I’m incredibly proud of. I’ve been able to bring my own creative vision to life in a way that I never could have before.
And that’s the real power of AI-driven design. It’s not about replacing human creativity. It’s about amplifying it.
The Road Ahead
We are at the very beginning of a new era in design. The tools are still in their infancy, and the possibilities are endless. It’s an exciting time to be a creator.
But it’s also a time to be thoughtful and intentional. It’s a time to think about the kind of future we want to create. Do we want a future where design is automated and generic? Or do we want a future where design is a collaboration between humans and machines, a future where technology is used to unlock our collective creativity?
I know which future I’m betting on. And I’m excited to see what we create together.
Frequently Asked Questions
How were these items selected?
Each item on this list comes from direct experience, either from building my own companies or from patterns I've observed across the 200+ startups I've invested in. I prioritize practical, actionable items over theoretical concepts.
Which item on this list has the highest impact?
It depends on your stage and context, but in my experience, the items near the top of the list tend to have the broadest applicability. That said, sometimes the less obvious items create the biggest breakthroughs for specific situations.
Are these recommendations still relevant in 2026?
Absolutely. While specific tools and tactics change, the underlying principles remain consistent. I update my thinking regularly based on what I'm seeing in the market and across my portfolio companies.