Overview

AI-powered design and creativity tools have exploded in the past couple of years, making it easier than ever to generate images, refine text, and even create entire branding systems. As someone who’s always looking for ways to streamline workflows and push creative boundaries, I’ve been experimenting with some of the top AI tools to see how they fit into my process.

While these tools offer some incredible advantages, they’re not without their flaws. Below, I’ll break down what I like and what I don’t about each of the AI tools I’ve been using lately: Adobe Firefly, MidJourney, ChatGPT, Fontjoy, and DALL·E.

Adobe Firefly: AI-powered image generation

What I Like:

Adobe Firefly is seamlessly integrated into the Adobe ecosystem, which is a huge plus for anyone already working in Photoshop, Illustrator, or Premiere. Unlike other AI image generators, Firefly is trained on Adobe Stock, meaning there are fewer issues with copyright and questionable licensing. I also love that it has tools specifically tailored for designers, like text-to-image, generative fill, and style transfer, making it easy to create customized, high-quality assets without leaving my workflow.

Another standout feature is vector recoloring—being able to instantly adjust colors in a complex vector illustration using AI speeds up my process significantly. Firefly also has an easy-to-use interface, which makes generating AI content feel like an extension of my design process rather than a separate, cumbersome tool.

What I Don’t Love:

One of Firefly’s biggest limitations is that it’s still slower and more restrictive than standalone AI generators like MidJourney. Because it prioritizes ethical AI sourcing, the results sometimes feel less creative or innovative compared to other AI models trained on a broader dataset.

Additionally, while Firefly’s generative fill tool is promising, it often struggles with blending AI-generated elements naturally into an existing image—which can make edits feel artificial if you’re not careful.

Midjourney: AI image generation

What I Like:

MidJourney consistently delivers some of the most visually striking AI-generated images I’ve seen. The textures, color palettes, and compositions feel much more artistic and dynamic compared to other generators. I love that MidJourney has a community-driven approach, where users can share prompts, get inspired, and refine their skills collaboratively.

One of my favorite use cases is using MidJourney for mood boards, concept art, and visual inspiration. It excels at dreamlike, painterly, and hyper-detailed imagery, making it an amazing tool for brainstorming new creative directions.

What I Don’t Love:

The Discord-based interface is one of my biggest frustrations with MidJourney. It feels unintuitive and clunky, especially when compared to more seamless tools like Firefly or DALL·E. I also find it frustrating that prompt control is limited, meaning small adjustments can lead to wildly different outputs, making consistency difficult.

Another drawback is that fine-tuning and editing AI-generated images is challenging. Unlike Adobe Firefly or DALL·E, MidJourney doesn’t allow for direct modifying/removing/replacing elements or minor adjustments, which means you often have to regenerate an image completely if something isn’t quite right.

ChatGPT: AI-powered writing & idea generation

What I Like:

ChatGPT has been a game-changer for me when it comes to brainstorming, organizing thoughts, and speeding up content creation. Whether I need help structuring an article, summarizing complex topics, or drafting a quick piece of copy, it’s incredibly useful for jumpstarting ideas and cutting through creative blocks.

It’s also been great for technical explanations and problem-solving, especially when I’m working on understanding coding limitations, UX research, or structuring project plans. Having an AI that can instantly refine my thoughts into something structured and digestible saves me a ton of time.

What I Don’t Love:

While ChatGPT is great for rapid ideation, I still have to heavily edit what it produces to make sure it feels authentic and natural. Sometimes the tone can feel too generic, overly formal, or overly enthusiastic, which doesn’t always align with my writing style.

Another drawback is that ChatGPT can occasionally hallucinate information, meaning it confidently provides incorrect or outdated answers. This is especially noticeable when asking about niche design topics, software updates, or coding best practices. I always double-check its responses before relying on them.

Fontjoy: AI-powered font pairing

What I Like:

Fontjoy is a simple but surprisingly effective tool for helping me find the right font pairings for a project. It’s great for quickly generating typography combinations without spending hours scrolling through Google Fonts or Adobe Fonts.

I like that it suggests both contrasting and complementary fonts, which helps me balance readability and aesthetics when working on branding, UI, or web design projects. The “Generate” button makes it easy to quickly cycle through different combinations until I find one that feels right.

What I Don’t Love:

The biggest downside of Fontjoy is that the control is limited. While it’s good at generating font pairs, it doesn’t always take into account practical design constraints, such as licensing, accessibility, or readability for certain use cases.

Also, the AI sometimes suggests fonts that don’t work well together in real-world applications, making it more of a starting point than a final solution. I often use it as a reference but still fine-tune typography choices manually to ensure they align with the project’s needs.

DALL·E: AI-powered image & editing capabilities

What I Like:

DALL·E is one of my favorite AI tools for creating and editing images with precision. Unlike MidJourney, it has direct inpainting capabilities, allowing me to modify specific parts of an image rather than having to regenerate the entire thing. This is particularly useful for removing objects, replacing elements, or blending AI-generated content into existing designs.

It also has a more natural integration with ChatGPT, making it easy to generate visuals based on structured prompts. The realism and adaptability of its outputs make it a strong tool for generating product mockups, concept visuals, and quick ideation.

What I Don’t Love:

While DALL·E’s editing tools are useful, the outputs can sometimes feel less artistic and more AI-generated, meaning they lack the painterly or highly-stylized aesthetics of MidJourney.

Another issue is that DALL·E struggles with fine details, especially when it comes to hands, text, and complex compositions. While it has improved, I still find myself needing to clean up AI-generated images manually when using them in professional projects.

Final thoughts

Experimenting with AI tools has been both exciting and challenging. Each of these tools have strengths and weaknesses, and I’ve found that combining them strategically leads to the best results.

• For ideation and artistic inspiration, MidJourney is my go-to, despite its frustrating UI.

• For editing and practical image generation, DALL·E offers better flexibility.

• For workflow integration, Adobe Firefly works well within existing design tools.

• For brainstorming and content creation, ChatGPT speeds things up but still requires human editing.

• For typography pairing, Fontjoy is a great starting point, but I never rely on it completely.

AI is powerful, but it’s not a replacement for creativity. It’s best used as a tool to enhance, speed up, and refine ideas—not to replace human decision-making. That said, I’m excited to see how these tools continue to evolve and improve.

What AI tools have you been experimenting with? I’d love to hear how they fit into your creative workflow! 🚀