AI is no longer a fringe experiment: it’s a mainstream mandate. But with that shift comes a new kind of pressure: to act quickly, to appear modern, to bolt on something “intelligent” before someone else does.
For many teams, this leads to reactive choices. Features get prioritized because they sound impressive, not because they solve a real user problem. Familiar interfaces get copied instead of questioned. And in the race to keep up, the signal of what actually matters to users gets lost.
The good news is: the path forward isn’t just about being faster or flashier. It’s about designing with care, context, and curiosity. This manual is your edge in the age of AI—so let’s dive in.
Table of Contents
When you think of technologies reshaping processes, you might imagine a renovation from the ground up, an act of tearing out walls and leveling buildings. After all, it’s easier to install new machinery if you clear everything out first, right?
The leap in technological advances over the past few years might tempt you and your design team to think from a contractor’s perspective, with questions like:
What do you need to raze, replace, or reinvent to accommodate this future vision?
Do you even have bandwidth for a teardown?
If you don’t have bandwidth, what AI can you add on to stay current and appeal to new buyers?
But designers aren’t contractors. We don’t demolish functional spaces that serve customers. We’re more like urban planners, learning how people work and planning upgrades that fit into the patterns of the cityscape.
In city planning, desire lines are the worn paths people make through grass when no sidewalk exists—they reveal the routes people actually want to take. In product design, your desire lines are the real workflows, habits, and shortcuts your users already take, even if they’re not what you originally built for.
💡 Hot tips:
Actively seek customer feedback, even if it’s negative
Plan and adapt based on real user signals, not assumptions
Observe how people use your product vs. how you imagined they would
What does it mean to be “AI-native”? This buzzword has been making the rounds over the past year as more companies adopt AI-focused roadmaps and marketing campaigns, so let’s break it down.
“Native” has been a term in the tech sphere for a decade now, no doubt borrowed from the definition of “native plants.” It’s now notably part of the phrase “cloud-native”, whose most-cited definition includes the words resilient, manageable, and sustainable.
If we extend this metaphor to AI, to be “AI-native” means building features and workflows that fit naturally into your product’s ecosystem and scale alongside your users by:
Blending seamlessly into your platform without feeling out of place or disjointed
Enhancing user experiences instead of harming or competing with them
Remaining resilient as your product grows and evolves
With this in mind, it’s unsurprising that AI agents like Spotter have had a meteoric rise in the past few months, precisely because of their ability to adapt and scale within an infinite number of landscapes.
Answering these questions will set your team on a path to building AI products people love and can’t wait to use. Just remember: native over novelty, any day.
💡 Hot tips:
Will the features be adaptable to additional features, workflows, and user growth?
Are they symbiotic with the rest of the platform, amplifying overall capabilities?
What do your customers think of the plans? (User research has entered the chat again!)
To successfully adapt to the changing landscape of our industry, it’s critical to stay agile and innovative. But let’s be real: agile is an exercise of the mind as much as one of process, and AI innovation doesn’t happen in a vacuum.
The best ideas are built on exposure, borrowing, and leveraging what already exists. Most major consumer tech companies are prime examples of this:
Uber commoditized a reliability gap in taxis
DoorDash tackled restaurant deliveries by rethinking takeout coordination
Amazon revolutionized package logistics through new distribution networks
Any technological impact AI has on the above process is iterative. None of those companies conceptualized the first idea of their services, but by embracing new technology and ways of thinking, they were able to address consumer pain points and disrupt existing markets.
You can leverage the same mindset to stay ahead. Remember, there are always unsolved product gaps or processes and workflows that can be improved, and it’s your job to find them.
💡 Hot tips:
Focus your lens on products in your vertical that are incorporating AI
Study AI-built products unrelated to your vertical to learn from their approach
Use an AI chat platform to collate research on emerging user demands
No matter how much you know about AI, it’s still a new technological relationship you have to cultivate the same way you learn a new skill. Familiarity and comfort take time, exposure, and repetition.
For example, have you used Google or talked to a product support chatbot online? If you said yes, you’ve likely had incidental interactions with artificial intelligence. But instead of simply searching for a topic or starting a return, go deeper and turn your attention to the process.
Rather than just engaging with your default day-to-day tools to answer questions, why not open a generative AI platform and see what comes out?
Ask questions like:
What are the best sources of news on [X] topic?
What are the return policies for [Y] company?
How can I optimize [my least favorite process]?
Consider factors like the way the interaction started and ended, the language used by the AI, and whether it surfaced helpful or frustrating responses. Did it give you the output you were looking for?
It may seem like a hurdle to stay curious and updated, especially if exploring AI is a mandate from leadership. But you can approach it as a small, intentional step further in your current day-to-day technology exposure.
💡 Hot tips:
Listen to curated news headlines about AI while you’re listening to the news
Watch videos about the hottest new AI tools while you’re taking a break
Collect AI design inspiration as you go by bookmarking features and ideas to try
For product and design teams with a directive from leadership, it’s easy to get trapped designing SaaS products and AI systems to solve problems users don’t have. When generic chat interfaces are most people’s baseline for what it looks like to interact with AI, it’s all too easy to copy the UI or chase feature parity for things that aren’t relevant to your customer base.
After all, innovation doesn’t come simply from iterating on a competitor’s UI or copying it. Let me tell you a story about 3M Command Strips, one of the stickiest products ever (pun intended). This anecdote was shared with me early on in my product design career, and has remained deeply ingrained in my approach to solving customer needs.
In the wall hook market, companies initially competed with each other, selling products that made creating holes in walls an easier task by:
Distributing weight differently
Developing custom anchor technology
Changing the angle of the metal nail relative to the wall
But in the mid-nineties, 3M realized the error in underlying assumptions about user needs—average customers didn’t care about the hanging method, they just wanted their stuff on the wall.
With this realization, Command Hooks generated $10M in revenue within their first 3 years of product launch. According to a market research report, as of 2024, the Global Command Hooks market was valued at a whopping $3.1 billion.
This is an amazing example of how lucrative it can be when you refocus on user needs, even in light of a “solved” user problem.
💡 Hot tips:
Tailor your problem to your user base, not your competitors’ exact approach
Validate user needs before committing resources to new AI features
Look for opportunities to remove friction entirely, not just improve the existing process
As the AI landscape changes rapidly and unexpectedly, you can choose to be afraid and resistant, or be pragmatic and start mapping out the future you want to build. If you follow the steps above, you will be among a very small percentage of the world population molding the future of AI and product design with intention.
ThoughtSpot helps you scale your product with intention. And that means designing and using AI as a multiplier, rather than a replacement, of the people you build for. Interested in test-driving the latest AI innovations yourself? Start your free trial.
1. What makes AI design different from traditional product UX?
Designing for AI means you're no longer shaping screens—you’re shaping intelligence. Interfaces must feel collaborative, transparent, and responsive, not just flashy. The goal is to embed AI seamlessly into user workflows, not bolt it on as a feature.
2. What does “AI-native” design actually mean?
“AI-native” refers to features and workflows that are built to integrate seamlessly with AI from the start—not added as an afterthought. It’s about enhancing user experience without disrupting it.
3. How do I balance AI innovation with user familiarity?
Start by understanding your users’ existing workflows (your “desire lines”). Then experiment with features that reduce friction, not just those that sound innovative. Test iteratively and focus on real needs over novelty.
4. Can AI replace the product design process?
AI excels at speeding up ideation and handling lower-level tasks, but a human-in-the-loop approach is still essential for strategic, intentional, and empathic decision-making.
5. How do I know if my AI feature is solving a real problem?
Ask: Would this be valuable even if AI weren’t trendy? Validate through interviews, usability tests, and usage analytics, not assumptions or competitor mimicry.
