How to approach AI?

“The best solution can usually be found in the best definition of the problem” - Chip Kidd

For every headline celebrating AI's capabilities, there's another warning about its limitations. But here's a simpler truth: AI works best not as an oracle delivering answers, but as a diligent partner offering possibilities.

Making AI work for you

AI can be thought of as a bright but sometimes inexperienced colleague. While it can process vast amounts of information and spot patterns we might miss, it lacks the nuanced judgment that comes from real-world experience. It's not here to make decisions for us – it's here to help us make better ones.

This section explores how to build a productive relationship with AI tools. We'll examine why treating AI as a suggestion engine rather than a source of truth leads to better outcomes. You'll discover how AI excels at jumpstarting projects and polishing final drafts, while learning why the creative middle ground remains distinctly human. Most importantly, you'll understand how to harness AI's pattern-matching power while keeping your own expertise firmly in the driver's seat.

• AI is a Suggestion, Not Direction

• Gen AI Helps with Hot Starts and Refinement

• Prompt Engineering is Coding in your Natural Language

• Treat AI as a Thought Partner

• AI first Products will work differently

AI is a suggestion, not direction

Al helps improve on your average

Amplifying human judgement, not replacing:

• AI doesn't tell you what to do. Consider, don't follow blindly

• AI is a forklift for the mind - it helps us “lift” more

• We encounter this frequently in daily life: fastest routes, what to watch and smart replies

Content recommendations that you might like are educated guesses, not commands. This same principle applies to all AI tools - they're suggestions based on patterns, not prophecies.

AI helps us process vast amounts of information and identify patterns that might take hours to spot on our own. When we're overwhelmed with data or decisions, AI tools can offer shortcuts and insights - but they don't make the final call. That responsibility stays firmly in human hands.

Consider how we use navigation apps. They suggest three routes, but local knowledge might tell you otherwise. Maybe you know about that school letting out or that hidden shortcut. This is precisely how AI works best - offering options while letting human experience take the wheel.

The real power of AI emerges when we're stretched thin. Think of AI as a talented associate who brings fresh perspectives to your work. When you're writing emails, it might suggest responses. When you're shopping, it highlights products you might like. When you're choosing movies, it presents options based on your viewing history.

But just as you wouldn't let an associate make all your decisions, you shouldn't surrender your judgment to AI.

Gen AI helps with starting and refinement

The middle ground is messy and best done by humans

The sweet spot for AI assistance is beginnings and endings:

• AI gives a hot start - brainstorming, assisting with idea generation, etc

• AI is good at reviewing, proofing, and using as a cross check

• AI doesn't deal well with the messy middle well - that is the human part

When a writer is facing a blank page, a looming deadline or polishing a final draft, AI can help. Tools like AI can help transform the creative process at its edges while leaving the core uniquely human.

At the start, AI excels as a boundless brainstorming partner. It generates initial concepts, visualizes data patterns, and creates scaffolding for your work. At the other end, AI becomes your meticulous reviewer. It proofreads with precision, checks cross-browser compatibility, and ensures regulatory compliance.

But between these bookends lies the messy middle – where human insight is hard to replace. Here's where we select the most promising ideas, craft meaningful experiences, and understand what truly matters to users. This creative heartland demands human judgment, empathy, and strategic thinking that AI cannot replicate.

The future of work isn't about AI replacing humans – it's about recognizing where each partner shines. AI handles the heavy lifting at the start and finish, while humans navigate the crucial journey between.

Prompt Engineering is coding in your natural language

Best solution is found in the best definition of the problem

Code with everyday words:

• “The best solution can usually be found in the best definition of the problem” - Chip Kidd

• Prompting skills are the same as framing problems and defining what success looks like

• Many different ways of framing to consider… from RISEN to COSTAR and more

What if you could code using everyday words? That's prompt engineering – where natural language becomes your programming tool.

On one side stands human intention; on the other, AI capability. Your prompts act as a bridge, turning vague wishes into clear instructions. Just as a well-written function tells a computer exactly what to do, a well-crafted prompt guides AI toward your desired outcome.

The secret lies in problem definition. By clearly stating what you want, why you want it, and what success looks like, you shape AI's response. Each word in your prompt acts like a line of code, steering the AI's processing power toward your goal.

Frameworks like RISEN (Role, Input, Steps, Example, Next steps) offer structure to this conversation. But beyond frameworks, prompt engineering draws on timeless skills: clear communication, precise language, and systematic thinking.

What makes a great prompt engineer? The same qualities that make a great problem solver: curiosity about how things work, attention to detail, and the ability to break complex tasks into simple steps.

You're not asking AI for help – you're teaching it how to help you.

Kanika B K

uxGPT: Mastering AI Assistants for User Experience Designers and Product Managers

RECOMMENDED REsource

An essential read with practical strategies to harness AI Assistants to plan and brainstorm user experience and product management activities. By mastering these prompts within the design thinking process, you'll unlock new ways to streamline workflows and generate innovative solutions.

Treat AI as a thought partner

Ask for ideas, not answers

AI is best equipped to give you ideas, feedback, and other things to consider:

• Give AI enough context to make associations

• Ask AI to run your problems through decision framing - and have a conversation

• Share your data and see from a different lens

When you approach AI systems with curiosity rather than seeking absolute answers, you unlock their real potential.

Consider how you brainstorm with colleagues. You share context, bounce ideas around, and challenge assumptions. AI works best in exactly this way. Instead of asking "What's the solution?" try "What angles haven't I considered?" or "How might we approach this differently?"

The key lies in rich context. Just as a friend can't give meaningful advice without understanding your situation, AI needs background to make valuable connections. Share your constraints, goals, and previous attempts. Then watch as it draws unexpected parallels and surfaces hidden possibilities.

Make it defend its thinking. Push back on suggestions, ask for clarification, and explore trade-offs. This back-and-forth reveals both the strengths and limits of its understanding while deepening your own insights.

Your expertise remains central. AI won't replace your judgment – it amplifies it. Use it to pressure-test ideas, generate alternatives, and challenge your assumptions. Think of AI as a tireless collaborator ready to explore countless possibilities while you steer toward what truly matters.

AI first products will work differently

Native AI workflows will likely shift what tools look like and how we use them

Beyond a blank canvas:

• Websites and apps started off originally as digital magazines

• AI tools will move beyond today’s tools with a chatbot attached

• Answer and solution engines instead of search engines

Websites started off as digital magazines. We've moved far beyond those early attempts to squeeze print into pixels. The same transformation is happening with AI-powered products.

Tomorrow's AI tools won't simply be today's software with a chatbot attached. They're evolving into something fundamentally different. Consider how an AI-native product works: instead of starting with a blank page, you begin with suggestions. Rather than searching through endless options, you explore guided possibilities. The system learns and adapts with each interaction.

These tools embrace iteration, turning rough ideas into polished work through continuous refinement. They seamlessly blend text, images, and code – what we call multimodal interaction. They excel at remixing existing content into fresh combinations while maintaining context and meaning.

Think of it like the shift from libraries to search engines. Libraries organize existing knowledge. Search engines made that knowledge discoverable in new ways. Now, AI-native products are becoming answer engines, actively helping create solutions rather than just finding them.

What does AI mean?

The digital age brought us information. The AI age brings us choices – countless decisions about how to harness this technology that seems to evolve daily. From doctors analyzing medical scans to writers crafting stories, AI reshapes not just what we can do, but how we do it.

About AI Demystifying

You don’t have to be an expert to understand AI, just like you don’t have to be a mechanic to drive a car.

But it can be challenging to sort through the noise - and we need cartoons in our heads about how technologies work.

AI Demystifying is a place to begin sorting through the hype, unpacking foundational concepts and developing frames of reference for AI.

Process is a set of tools, not rules.

AI Demystifying is another UX How Tool from Method Toolkit LLC.

Explore More like AI Demystifying from UX How

CoDesign AI

Collaborating with AI and each other in building experiences.

UX Designer Guide

UX and Product Designer insights for navigating design realities.

A collection of prompt engineering techniques for UX.

XD Prompts

About UX How and T. Parke

UX How is a set of UX & Product Design “How To” sites with insights, resources, and blueprints for Design, UX and AI.

T. Parke is the Director of UX How with prior experience at ESPN, Disney, and Alaska Airlines. He has previously been a design leader on projects for Rolling Stone, Microsoft, Nickelodeon, and Marvel.

There you go.