AI Prompts To Augment Your Design Workflow & Generate New Ideas
8 Principles & Prompts for Strategic Human-AI Collaboration
You've probably heard this phrase echoing across social media lately (and I’m guilty of repeating it too):
"AI won't replace us — but a designer using AI will."
Here’s the thing: as a Senior Product Designer working on AI features at Google, I’m starting to believe in the power of human-AI collaboration. While some of you might consider this a buzzword, I believe it’s a fundamental shift in how we work. The real strength isn't in mastering every new AI tool in isolation, or writing perfect prompts on the fly. It's in the art of combining our deep user empathy, strategic thinking, and unique problem-solving voice with AI as a true collaborator and an amplifier of our abilities. Viewing AI as a replacement for our essential human contribution, I think, is shortsighted and overall dismissive of the transformative partnership we have at our disposal.
Instead of adopting this attitude, I have instead shifted to focusing on the areas that AI is not very good at, the areas where our human abilities differentiate us. My goal has been to upskill those areas to build what I’ve been referring to as a creative moat. The way I envision this moat is as a blend of our strengths (contextual understanding of user needs, strategic decision-making, and nuanced understanding of human behavior) augmented by AI when used intentionally and thoughtfully.
Today, the way that most people are engaging with AI is through Large Language Models. While these are easy to engage with via natural language, there are patterns and methods of engaging with these models that can make the results even more powerful. This method of communicating effectively with LLMs is dubbed “prompt engineering.” I think there is hidden potential in effectively communicating with LLMs through prompt engineering, especially from the perspective of UX design. I kept wondering, if we stress-tested this hypothesis of leveraging AI’s strengths, what use cases and prompts would be best to engage AI with? Would it be possible to create a set of guiding principles or prompts that we could systematize for different use cases?
This toolkit is an MVP attempt at defining an initial set of 8 guiding principles, each accompanied by practical AI prompts, designed specifically to help UX and Product Designers master collaboration with AI. It's important to clarify: these principles and prompts themselves are not the complex "AI Workflows" or "Agentic Patterns" that leading AI researchers discuss. Instead, our principles are the strategic guidance, and the prompts are the individual building blocks, for you to initiate and participate in these more sophisticated AI workflows. This practical, workflow-centric approach aims to enable us to:
Uncover hidden user needs and identify market opportunities far earlier.
Make smarter, more data-informed, and profoundly human-centered design decisions.
Navigate uncertainty and adapt strategically throughout the product development lifecycle.
Protect and significantly grow our unique value as design leaders and innovators.
Think of these principles as our essential AI engagement tools, ready to be deployed whenever we're charting new territory (whether it's kicking off a fresh product feature, wrestling with ambiguous user research findings, or pushing the boundaries of our design solutions to create something truly impactful).
How to Get the Most Out of This Toolkit
Keep in mind, this isn't meant to be a rigid rulebook; it's a practical resource built for immediate application you can modify based on your needs and use cases.
A Note on Experimentation: The prompts you'll find here are a first draft born from my own experimentation. I've designed them to be as helpful as possible for our roles as designers, but the best way to truly understand their value is to try them out yourself!
Start Anywhere: Pick the principle that speaks most to your current challenge or curiosity. There's no fixed order or "must-do" sequence.
Use the Prompts as Jumpstarts: Each principle includes a detailed, copy-paste AI prompt. These are designed to be dropped into your AI tool of choice (like ChatGPT, Gemini, Claude, etc.) and customized with your specific project, user, or product context.
Go Deeper with Questions: Use the strategic questions to broaden your thinking, uncover overlooked angles, and apply the AI's insights more profoundly.
Adapt and Iterate Relentlessly: The prompts are templates, not commandments. Modify them to fit your unique style, project needs, specific user personas, or any stage of your product's lifecycle.
Document Your Discoveries: Every insight you gain, every problem you solve with AI's help, is a brick in your creative moat. Make sure to capture it somewhere for future reference.
Share Your Results! After you try these prompts for your own use cases, share your results in the comments or in our Substack chat space. I will create a new thread dedicated to testing these prompts. Learning from each other's experiences is how we'll all grow our creative moats together.
Each principle is broken down into four parts:
What it means: A clear explanation of the principle and its direct relevance to UX/Product Designers.
Role for AI: How to instruct the AI to act as an "expert" collaborator for this specific task.
Copy-Paste Prompt: The ready-to-use, customizable prompt.
Exploratory Questions: Thought-provoking questions to deepen our analysis.
How to Use This: Practical instructions for applying the principle.
With all of that out of the way, let’s dive into the principles and prompts!