Transforming UX with Generative AI
From linear journeys to tailored terrain
Last June, Jakob Nielsen published an article that really caught my attention: “AI: First New UI Paradigm in 60 Years.” In it, he talked about a major shift in how we interact with technology. Instead of giving computers step-by-step instructions like we’ve done for decades, we’re moving to a new era of intent-based interactions. This means that technology will understand our goals and help us achieve them without needing detailed instructions.
As a UX designer, this shift can be hard to grasp. What does it mean for our work? Nielsen’s insights point to a future where user experiences are dynamic and hyper-personalized, adapting in real-time to each user’s needs.
This article is a deep dive into the emerging landscape of GenAI-first UX. Imagine transitioning from rigid, linear user flows to flexible, intuitive experiences. We’re exploring how technology can adapt to us, making interactions more human and responsive. This matters because it’s about making technology work for us, seamlessly blending with our human aspirations, and transforming digital experiences into something truly personalized and user-centric.
(Hyper)Personalisation
In today’s rapidly evolving digital ecosystem, a compelling narrative underscores the importance of UX: the shift towards deep personalization. Modern users no longer resonate with generic interfaces; they seek experiences woven around their unique preferences and behaviors. It’s hard to pinpoint where this desire came from, but it’s not a stretch to say it has been influenced by content algorithms prevalent in news, social media, and e-commerce.
If algorithms can tailor content recommendations, why shouldn’t they shape our user journeys, interfaces, and the entire digital experience?
Unpacking personalisation
At its core, personalization in UX means adapting various platform elements to cater to individual user nuances. This can manifest in several ways:
- Content: Personalized text or media, like using a user’s name in greetings or adjusting articles based on age or cultural background.
- Tone: Adopting a formal tone for professional users while switching to a casual tone for younger users.
- User Journeys: Customizing the sequence or pathway a user follows based on their objectives and familiarity with the platform.
- Layouts: Adapting visual arrangements to user preferences or devices, such as grid-based or list-based layouts, dark mode, or light mode.
Why personalisation matters
Personalization in UX design is crucial for engaging users and driving loyalty. Users who feel understood and receive content tailored to their preferences are more likely to interact with and stay loyal to a platform. This leads to increased conversions, higher customer satisfaction, and a stronger bottom line. To really see how powerful personalization can be, take a look at these compelling statistics.
Designing for the audience of one
Most digital experiences today rely on segmentation, slapping labels like “millennial techie” or “soccer mom” on us based on often vague demographics and psychographics. Some platforms, like social media, try to refine this with “micro-segmentation,” but it still feels like we’re being fit into pre-made boxes.
The future, however, is far more exciting: hyper-personalization. This approach throws out the boxes altogether and focuses on unique experiences for every individual.
Imagine this: you need new running shoes, and your phone recommends a pair with the perfect level of support based on your recent workouts. No more wading through generic options — just the perfect fit for you at that specific moment.
Hyper-personalization doesn’t just stop at recommendations. With Generative AI and enough context, technology can finally cater to the spectrum of our needs. We’re not one-dimensional beings; our preferences, interests, and needs shift depending on the context. Hyper-personalization recognizes this and adapts accordingly.
Ultimately, it’s about making sure technology serves each of us as individuals, not just fitting us into a box. Designing for the audience of one means that every user feels truly understood and valued.
AI as an enabler
With the growing desire for hyper-personalized experiences, the question isn’t “if” these experiences will happen, but “how?” AI is the key to unlocking infinite customization and configuration. We no longer need to translate our intentions into specific commands. Instead, we express what we want, and the AI handles the “how.”
Why is this better?
Let’s make an example. Imagine you need to cite a paper. In the old paradigm, you’d break down the task of choosing a style, learning its rules, gathering information, and formatting each citation correctly. It is time-consuming and detailed work! In the new paradigm, you express your intent — “cite the sources in this paper using Harvard style.” The AI translates this into the necessary steps, executes them, and presents the results for your review. You focus on your goals, not the intricacies.
This experience is better because it’s more human-centric. You make requests naturally, like you would to a helpful colleague. This humanized interaction will open the digital world to everyone, from helping you find the perfect running shoes to effortlessly citing sources in your next paper.
AI doing its thing
Now that we’ve entered the new paradigm of intent-based interactions, traditional software tasks that required step-by-step navigation are becoming obsolete.
If AI is the invisible engine powering our hyper-personalized experiences, how do we design for this silent partner?
Silent AI processes are a double-edged sword for UX design. Features like real-time summarization or AI-powered trip planning offer incredible value, but their background nature can leave users confused. Traditional tricks like progress bars are often insufficient, especially for complex tasks handled by AI agents like booking a dream vacation.
The key lies in crafting new design patterns, prioritizing transparency, and building trust. Think of it like designing for a non-technical user. They don’t need to understand the complex algorithms, but a basic sense of the process and potential timeframe is crucial. Instead of generic loading messages, sprinkle subtle cues throughout the user journey. This could involve micro-animations that depict AI activity, contextual messages like “Your AI assistant is crafting the perfect itinerary!” or calming visuals that signify background processing is underway.
By strategically implementing these new transparency measures, we can ensure a seamless user experience even when the powerful force of AI is working diligently behind the scenes.
The experimentation phase
We often take familiar UX patterns like Infinite Scroll for granted, forgetting the countless iterations and experimentation that birthed them. The dominant interaction patterns for GenAI experiences remain in discovery, but one thing’s certain: we won’t find them without embracing experimentation.
Weird, then good
Exploration is crucial in UX design, even if it initially leads to results that may seem unusual or unconventional. Many of our early attempts may appear odd or even comical in hindsight. Still, through persistent iteration and collective effort, we can uncover designs that truly resonate and stand the test of time.
To understand the current state of GenAI design, let’s take a trip down memory lane with mobile calculators. The Wayback Machine transports us back to 2011 when a handful of popular mobile calculator apps showcased a potpourri of design approaches that might make you chuckle and wonder, “What were we thinking?” We witnessed designers grappling with visual elements and new ways to interact with these digital tools. There’s a clear tension between mimicking real-world experiences (realism) and taking advantage of the digital world's unique possibilities (affordances).
Fast-forward two years: Gone are the skeuomorphic shadows and clunky buttons, replaced by a minimal, flat design integrated (in this example) with a novel swipeable drawer for currency conversion. This begs the question: Would users have intuitively understood the swiping functionality from the start? Or was the journey through various design iterations crucial for us users?
The pull-to-refresh moment
David Haong summarises the situation clearly when he says, we are waiting for AI’s pull-to-refresh moment — a defining user interaction that will set the standard for future interfaces. As we transition into the AI era, the challenge is to explore beyond conventional designs, awaiting that transformative idea.
A whole world of new interactions and UX patterns is waiting to be discovered.
What will GenAI’s ‘pull-to-refresh’ moment be? We will only know through experimentation, collaboration, and conversation.
From platforms to open worlds
It’s hard to place this industry moment with a comparative opportunity, but one that springs to mind is the moment when video games moved from platformers (2D side-to-side movement) to open-world (3D freedom of movement) games. While much of the game logic and narratives are portable, there is a noticeable change in user agency. GenAI allows us to make the same jump, but we must ensure we provide the right amount of constraints to keep users safe while encouraging them to enjoy the service or complete tasks using their desired approach.
Constraints as an agent of storytelling
If you have ever set foot on a sports field, you will notice all sorts of strange markings on the grass. These are just one level of constraints to the players and they help shape the game, which helps narratives form and fan bases to unite around a team and a sport. We also see constraints outside of the world of sport, think of a director masterfully framing a shot for a film to fit a 16:9 screen or a fulfilment specialist changing an approach to packaging and helping get more products out of the factory in a more efficient manner.
We humans seek constraints because they help us make sense of situations and begin adapting.
As designers and product experts, we have a responsibility to build constraints into our experiences to give them context. Traditionally, we have done this by limiting the number of pathways available to individuals at any given time and we normally picture this as a nifty tree diagram. The challenge here is that interacting with an AI doesn’t have an inherent path and each individual’s objective and context will drive their flow. So we need to define certain rules of our system, almost like walls in an open-world game, that define what’s not possible, without ascribing a finite list of possibilities. This will allow the user to feel free without our system while maintaining their safety and our brand communication at all times.
Open world UX
One way we might describe the systems we are discussing would be Open World UX. Open World UX has several distinguishable features, they are predictable and consistent, they afford clear narratives and they allow for tailored experiences. Open World UX experiences need to ensure that for each user input or action, there’s a consistent and expected response, with variation (read accuracy) fine-tuned based on the scenario. It’s like an artist who knows precisely which colour will emerge when two paints mix, but who in certain scenarios is open to nuanced surprises. Open World UX experiences also create clear narratives, ensuring that the user’s journey has clarity, purpose, and direction. Finally, these experiences can be fine-tuned by designers by adjusting the constraint set, like a composer adding or removing instruments to the orchestra and therefore affecting what can be played and how.
For designers and product experts, these open-world experiences create opportunities for truly custom interactions, adaptive layouts, and immediate feedback loops. Designers can curate specific interactions for users, ensuring that each touchpoint feels personalised. Given the defined constraints, the design can dynamically adapt to user preferences. Finally, any changes or enhancements in design can be instantly reflected, allowing for real-time (non-predetermined) iterations based on user feedback.
From linear to dynamic journeys
The realm of UX has seen many changes, but linear user journeys remain foundational. These journeys guide users step-by-step towards a predefined goal, based on user research and business objectives. However, organizations often miss opportunities to serve users at the long tail of their customer base, deeming it not worth the investment. With Generative AI as an orchestrator, designers can now create dynamic journeys that adapt to each user’s context and objectives. By using modular approaches in our services, we provide users with tailored experiences, ensuring they receive exactly what they need.
Linear journeys
In the early days of digital interfaces, many users found digital interactions confusing. Linear flows provided an easily understandable structure for both users and machines, especially for sequential tasks like setting up a device or finalizing an online purchase. Their strengths lie in simplicity, guidance, predictability, and ease of implementation. Clear, straightforward paths eliminate confusion and help users complete their goals. Additionally, linear journeys allow users to easily predict what will happen next. These clear, predictable steps can be easily translated into programmed functions, making them straightforward to build.
Constraints of linear journeys
True personalization, a growing necessity in modern UX, struggles within the constraints of linear designs. Linear journeys inherently have difficulty catering to individualized preferences, behaviors, and contexts. For example, a music app using a linear approach might guide a user through a series of steps to create a playlist (create, name, add tracks, etc.). In contrast, a dynamic model would streamline this into a single step (create playlist), considering user preferences and external factors like trends.
Moving to the meta-level
Today, the digital environment demands a more adaptive and individualized approach.
Linear journeys should be viewed as the meta-level of user interaction with our product or service.
For example, a linear journey for purchasing a new coat (browse, select item, purchase) can incorporate dynamic elements within those steps. Each user can browse in their preferred way, such as by voice, using recommendations, or following trends. Moving towards dynamic journeys, powered by AI, promises us the freedom to shape our own experiences.
Dynamic journeys
Dynamic user journeys are key to achieving hyper-personalised experiences. Journeys are coordinated by an AI that uses an individual’s context and objective to assemble a productive experience on the fly.
These journeys are hyper-personalised, objective-driven, and ephemeral.
When considering how this works, familiar questions arise around AI Doing Its Thing and the simplicity and predictability achieved by linear journeys. Here, we must achieve the appropriate balance of dynamism, wrapping our dynamic experiences inside meta-linear journeys.
An example
Imagine you would like to go to your nearest train station and you have only a paper map at your disposal. You would likely locate yourself and then consider what roads, or transport systems might take you to the train station. You may even draw your route on the map. Additionally, you would probably pay detailed attention to relevant street names so that you can orient as you go. In the end, you would have a set of directions that will take you to the train station, and then you would set off and follow your plan.
In contrast, what if you had a GPS? In that case, you would enter your destination of ‘Train Station’ and then review and select your travel options (public transport, walking, biking, etc.). Then you hit start. If the unexpected happens such as traffic or train delays the GPS immediately recalibrates in real time. Despite unexpected changes your focus, rather than on street names or subway stops, is on your destination. This is a dynamic journey.
Objective-driven experiences
Our dynamic user journeys are directly aligned with Nielsen’s assertion that we are in the era of Intent-Based Outcome Specification. Borrowing from game design, we can see each user’s journey as a series of objectives or ‘levels’ to achieve, with linear meta journeys serving as the overarching narrative. As users progress, the system adapts, providing the right tools, content, and pathways, much like a game where each level offers unique challenges tailored to the player’s skill.
This brings a new challenge for designers and product experts: helping users clearly define their goals while ensuring they understand the AI’s background processes. Creating a transparent and supportive environment is crucial, allowing users to communicate their needs easily and enabling the AI to meet those needs effectively. Innovative design solutions are needed to bridge the gap between user intentions and the complex processes the AI performs to fulfill them.
Summary
We believe we’ve reached a pivotal moment in user experience design, transforming UX with Generative AI. Moving from linear journeys to tailored terrain, we can now embrace individual journeys enabled by this innovative technology. No longer do we need to adapt to technology; technology will now adapt to us. This shift is about designing for the audience of one — the long tail — everyone.
Why does this matter? Because it makes our experiences more human. By adapting to each user’s unique preferences, behaviors, and contexts, technology becomes more intuitive and relatable. Imagine all your digital experiences working like a GPS: adapting to real-time changes and focusing on getting you to your destination rather than following a fixed path.
Linear journeys provide structure, but dynamic journeys respond to your unique needs, creating a more personalized and engaging experience.
This approach not only enhances user satisfaction but also makes technology more accessible and democratized. It breaks down barriers, allowing everyone to enjoy personalized, intuitive interactions, regardless of their technical expertise. By making technology more human and user-centric, we ensure that digital products are inclusive, empowering, and truly tailored to individual needs.
This article marks our first deep dive into the GenAI-first landscape. We’re just scratching the surface of what’s possible.
About the Authors:
Marc Seefelder is co-founder & Chief Creative Officer at MING Labs.
Miles Johnson is AI Experience Lead at MING Labs.
We specialize in creating and implementing human-centered GenAI-first experiences. Our mission is to make any human better at what they do with the help of AI, enhancing every interaction and elevating your business with custom GenAI solutions tailored just for you.
Thank you for reading!
