Creative Projects Using AI Image Generators

Are AI Tools Good Enough for Real-World Applications? As an early adopter of new technologies, I've consistently integrated AI into my projects. I'm excited to share some recent outcomes where I've brought AI-driven 3D generations into my VFX workflow. 🎨 Starting with Innovation: I kicked off the project by using OpenAI DALL-E 3 to generate innovative 2D graphics to replace the contents of a bus ad. The AI's ability to interpret and visualize creative prompts offered me a selection of visuals to choose from. 📦 AI Texturing: The same AI-generated images were transformed into a realistic package design, textured over 3D models which I crafted in Blender. This integration showcases how AI can enhance the visual touches in a project. 🐻 Text to 3D: For the central piece of the VFX, I created a 3D gummy bear using Genie from Luma AI purely from a text description. Although the initial texture wasn't perfect, I tweaked it in Blender to achieve a gelatin-like look, ensuring the 3D form was just right. 🎬 Animation and Composition: Adding bones and animating the scene in Blender, followed by compositing everything in After Effects, allowed me to focus more on the technical details and composition, which are crucial in VFX projects. 🤖 The AI Advantage: The integration of AI tools into VFX not only reduces the time needed to create complex visuals but also broadens the creative possibilities, especially for those with technical skills but without an artistic background. This doesn't replace the creator, but instead provides them with more powerful tools to enhance the concept. Have you tested AI tools for your creative processes? Let me know which tools you’ve you have found to work best and adopted into your workflow.

What if you could generate studio-quality visuals from simple prompts—in minutes? This weekend, I tested ChatGPT's new AI image generation. With minimal prompting, the AI quickly created images that typically take weeks of planning and thousands of dollars. Examples of incredibly simple prompts used: 1. Glowing Obvi Bottle: "Create a luxurious, cinematic ad image of a sleek 'Obvi Bottle' with condensation, cool blue and pink lighting, and a soft mist adding a refreshing glow." 2. Dreamlike Obvi Burn Box: "Can you use these reference images to create a vibrant and colorful scene for our Burn Box? Should feature bright pinks, purple lighting, luxurious ad quality, and soft magical mist around the product." What blew me away wasn't just the quality—it was the speed. Each image took roughly 2 minutes to generate. For context, similar professional shots would cost: - Thousands for the shoot - 1-2 weeks of planning - Multiple revisions with photographers - Location permits or studio time - Product shipping and setup As a brand that tests hundreds of ad variations weekly, the implications are staggering. Key actionable takeaways for brand builders:  ✅ Rapidly iterate and test dozens of visuals daily  ✅ Validate ideas instantly without physical prototypes  ✅ Launch high-quality visuals quickly and cost-effectively  ✅ Test multiple product bundles before manufacturing  ✅ Create season-specific content without new photoshoots Honestly, this could change our entire creative workflow. Instead of committing to one expensive creative direction, we can now explore 50 concepts before spending a single dollar on production. I believe we're 12 months away from completely rewriting how brands are built. The future belongs to brands that master rapid, iterative testing. The simpler your process, the faster you'll discover winning ideas. And now you can do it with a pre-seed budget to create Series A assets. How are you leveraging simple AI prompts to accelerate your brand's growth?

This might be the most slept-on way to use AI creative tools. It’s called Spaces by Freepik. Spaces is a node based canvas. So you can connect text, image, and video nodes together to build a visual AI workflow. For example, let’s say I wanna make the perfect image of Michael Jordan dunking over a Lamborghini. Before, I’d have to type a text prompt and manually generate 10 different versions, while tweaking the style description every time. But Spaces lets me set up a workflow with 10 different branches that automatically runs this process for me. You can see, I click once, and automatically have 10 different image outputs from all the best image models. When you rig up multiple of these chains together, you basically create an entirely automated system for generating visuals. All we have to do is drop in a single line of text from our script, and this workflow spits out 5 AI generated videos, in our desired style, that are all ready to go in just a few minutes. This type of workflow was super difficult to build before, but with Spaces it’s much much easier. And the beauty of these workflows is that once you build them, you can share a link so that any other users, like your editors, can redownload the entire workflow on their side, in a single click. I think these visual canvases are going to be the future of how people use AI creative tools. Because they make the process feel less like work and more like a video game. If you wanna try this out, check out Spaces on Freepik. Follow Kane K. for more AI creative tips. #ai #artificialintelligence #tech #technology #freepikpartner #design #marketing #aidesign #animation #motiongraphics #productmarketing

Adobe just gave us the creative partner we didn't know we needed And it’s been helping me amongst the Q4 and Christmas chaos. Black Friday sales, Christmas campaigns, darker nights, and let's be honest, our creativity is already running on fumes as we countdown to the holidays. But when I was at Adobe MAX last month, they shared something that I’ve been testing as we approach the end of year content madness. Adobe Express AI Assistant (beta) isn't just another AI tool trying to replace creativity. It's the creative partner that helps you save time whilst still bringing the creativity out in you. As one of Adobe’s leadership executives mentioned during the summit: “AI shouldn’t automate aesthetics — it should amplify artistry.” Here's how I’ve been trialling the AI Assistant in Adobe Express: Once you toggle on AI Assistant, you're not starting from scratch anymore. You can either work with Adobe's templates or upload your existing designs. The AI assistant suggests prompts based on what you upload, so when you’re lacking that creative spark, it's already thinking ahead for you. I tested it with a few edits this week: • Apply vintage film effects — because who doesn’t love nostalgia during the holidays • Generate a design to elevate an existing image - I need creative inspiration for our Christmas adverts round up! • Change backgrounds completely — from office to Santa’s workshop But this is what I love most: Instead of generic AI suggestions, it learns your brand aesthetic. You can: → Add or replace objects while keeping your style intact → Apply filters and palettes that stay on-brand → Resize and realign without losing quality → Generate copy that matches your tone The feature works through simple text prompts, so if you need a last-minute Christmas invitation, simply prompt it with "create a Christmas dinner invitation with red and gold sparkles” and watch the magic happen. You can easily prompt anything and edit anything! This isn't about AI doing the work for you. It's about AI assisting with your creative workflow, so you can save time whilst still controlling the creativity of your assets. Give it a try and let me know what you think! 🤎   #AD #AdobeAmbassadorsAtMAX

Building tools with n8n Over the past few months, I’ve built a handful of small tools to simplify my creative process. What began as small side experiments slowly turned into tools I now use almost every day. I’m kicking off a little series to share a few of them — starting with a Image-to-Image Generation tool. One of the biggest challenges with AI image generation is maintaining a consistent look and feel. This little tool solves that. n8n workflow + Chrome extension I built a small Chrome extension that sends any image directly into n8n through a webhook API. The image is analyzed using OpenAI’s vision model, which generates a new set of prompts through a dynamic system prompt. Those prompts are then passed to Google Gemini 2.5 Flash Image (Nano Banana) to create a set of new images in the same style and with the same look and feel. One of n8n’s biggest strengths is the ability to build fully custom workflows tailored to your exact needs. I design all my workflows with a modular approach, making them easy to reuse and combine across different automations. Swipe to see the result 👉

 🎨 Visual Thinking just got a major upgrade! Google just announced that Image Generation in Gemini app is now available for all ages in Workspace for Education (Fundamentals, Standard, and Plus). This is a game-changer for student creativity. Under 18 safety guardrails: ✅Can refine a generated image with follow-up prompts. ❌Cannot directly edit images or use uploaded images for image generation. ‼️Youth-focused safety testing and content filters are in place, but filters are not perfect. This is a great way to safely leverage AI to help students "show" what they know and visualize their thinking. Here are a few creative ways students could use this across the curriculum tomorrow: 🧪 Science & Engineering: "Generate a concept art image of a sustainable habitat on Mars using only local materials." (Visualizing abstract engineering challenges). 📚 English Language Arts: "Create a movie poster for Macbeth that focuses on the theme of guilt using a red and black color palette." (Analyzing theme and tone through visual media). 📜 Social Studies: "A busy marketplace in 14th-century Timbuktu based on the travel writings of Ibn Battuta." (Bringing primary source text descriptions to life). 🎨 Art & Design: Rapidly iterating on mood boards or composition ideas before moving to physical media. Admins: This feature is manageable in the Admin Console by OU now. How do you see generative imagery changing how students demonstrate understanding? I'd love to see others' ideas in the comments! Image generated by me with Gemini image generator 'Nano Banana'

Recently, I wanted to experiment with a new workflow that involved partnering with AI to create children's coloring books in just two weeks. Here's exactly how I did it (and what went hilariously wrong) ↓ 🎨 Creative Direction I established a creative direction and leveraged AI as a collaborator to test the quality, efficiency, and accuracy of various image generation tools and models for high-volume output formats, such as coloring books. 🧪 Learning Tested image generation on ChatGPT (GPT-Image via GPT-4o), NanoBanana (Gemini 2.5).  - Style and aesthetic varied drastically between the two platforms. - Being explicit with the prompt and hand drawn model pages made all the difference. - New branches with original prompting helped reduce errors, hallucinations, and dilution. - These probability tools did a pretty good job with anatomy but there was still a lot of bug fixing to adjust animal anatomy for accuracy. 👀 Outcome (POV) - Keep humans centric to the process. As the art director there is no limit to what you can create with GenAI tools if you k - Prompting is hard work. Spend the time to define use cases well so you build the right thing. Think about it in Layers. - When the tools fell short, I edited and directed myself, and it turned into a real creative advantage. 🫧 Success! High-quality, fun coloring books for kids that I can now produce in custom themes at scale, something that would have been nearly impossible to do manually within this timeframe. Keep Exploring! P.S. All books are available on Amazon (links in the last slide). Proceeds fund more creative experiments! #GenAI #AIDesign #CreativeAI #ProductDesign #Publishing #CreativeDirection #ColoringBooks

While people are busy creating their cartoon characters and having fun with the new OpenAI GPT-4o image generation tool, I decided to test it on something a bit different: engineering use cases. Can a creative image generation model support civil and infrastructure engineering? It turns out, yes, with the right guidance (although not quite there yet). I explored three practical applications: Sea Level Rise (SLR) Simulation Scenarios Climate adaptation planning often relies on GIS maps and simulations. GPT-4o can create illustrative views of how a coastline or neighborhood might change under different sea level rise scenarios. These visuals are not analytical models, but they’re helpful for community engagement, early design workshops, and raising awareness about climate impacts. Construction Staging and Phasing Visualizing site conditions across phases, before excavation, during substructure work, and at completion helps teams, clients, and the public understand project timelines. GPT-4o can quickly generate visual representations based on a short prompt for different stages. This can accelerate site planning, communication, and permitting workflows. Urban Revitalization and Streetscape Improvements Instead of relying on generic renderings, GPT-4o can instantly generate visuals for urban renewal concepts, such as adding green spaces, bike lanes, or pedestrian-friendly designs. It can complement site sketches or planning documents, helping planners and engineers quickly prototype ideas visually. Let’s be clear: AI doesn’t replace engineering expertise. These tools don’t understand structural design, drainage, or traffic volumes. However, early-stage communication, idea generation, and stakeholder alignment can significantly boost human engineers productivity and creativity. We are not being replaced, we are being augmented. #AI #GPT4o #CivilEngineering #UrbanDesign #ClimateAdaptation #ConstructionTech #AIDesignTools #OpenAI

🚀 Want to create standout Amazon main images in MINUTES using AI? Here’s how I do it: By leveraging AI tools like DataDive and MidJourney, you can quickly generate dozens of creative concepts for your product images. Start by running a product brief with DataDive—it scans all the reviews and detail pages for your top competitors, then breaks down what customers love, dislike, and even who your ideal buyers (avatars) are! Once you have insights—like must-have features or what makes your product unique—you can use MidJourney to create scenes and environments tailored to your top customer profiles. For example: show your product in the stylish home of a “busy professional” or space-saving setups for “apartment dweller Sarah.” You won’t upload these images to Amazon directly (they’re starting points), but it’s an incredible way to quickly brainstorm and communicate ideas to your creative team or agency. Think 30 prompts, 8+ usable image concepts—all in less than an hour. Embrace AI for faster, smarter Amazon image creation. Your listing (and your conversion rate) will thank you! To see this in more detail, check out our latest YouTube video (link in comments).