AI-driven UX curates familiarity, but in doing so, it narrows the path to new experiences.
Credit: B. Godart/Shutterstock
It’s Monday, and you’re on your way to work. Bracing for 45 minutes of bumper-to-bumper hell, you set your playlist to Commute. Maybe, like me, your music taste is rooted in nostalgia. In my case, that means ’90s and 2000s alternative rock — bands like Radiohead, Cake, and Queens of the Stone Age.
I use Spotify, which primarily employs advanced algorithms to recommend music based on my listening habits. Which unsurprisingly prescribes similar music to that of my benchmark alt-rock preferences. It’s an safe and effective method to keep listeners engaged.
But this predictable system of content delivery raises an important question — if my baseline music preferences were shaped by happenstance rather than algorithms, what does that imply about the role of algorithms in shaping my future tastes?
You Can’t Have Your Cake and Stream It Too
When I was fourteen, my aunt bought me my first music CD — Fashion Nugget by Cake. I hadn’t heard of them before, but she thought I might like them. My aunt was right — I loved their music, and I still do. I must have listened to that disc hundreds of times.
Cake had a distinct sound — deadpan vocals, funky bass, and trumpet accents — that stood out in the late ’90s. Their music felt effortless yet unique, mixing irony with solid rhythms.
Fashion Nugget by Cake | 1996
But times have changed. Nobody buys music for anyone anymore. Random opportunities to discover new artists died along with the surprise of answering the phone without knowing who’s calling.
Paradoxically, we now have more music at our fingertips than ever, yet organic discovery is disappearing. AI-driven platforms like Spotify are among the many digital products stripping the soul from life’s unpredictable moments — the kind that once led to unexpected joys.
So, what’s the point of this personal anecdote?
This story isn’t just about how music discovery has changed — it highlights a fundamental paradox in modern user experience (UX). In our quest for seamlessness, we may be eliminating the very friction that sparks new preferences.
The Evolution of Personalization — From Exploration to Curation
Every preference we hold today was once an unknown. Whether it’s a favorite music genre, a preferred app layout, or a go-to news source, these affinities took shape through unfiltered exploration or chance encounters.
In the beginning, streaming apps and digital platforms offer a broad mix of options. As we engage, algorithms gradually narrow the selection to match our tastes. While this personalization enhances convenience, it also limits exposure, reinforcing familiar choices rather than expanding our horizons.
Beyond entertainment, AI-driven curation shapes our social interactions as well. Social and dating apps steer us toward like-minded circles, creating curated bubbles where we primarily encounter content and people who align with our existing or assumed preferences.
In my own life, it’s possible that algorithm-driven dating apps would have filtered out my wife and me as a match due to our ten-year age gap, preventing us from ever meeting. But we met organically, beyond the reach of algorithmic matchmaking, and have been happily married for over 15 years.
The Need for Intentional Exploration
If UX designers want to preserve the richness of discovery, we must rethink how personalization is implemented. Rather than functioning as a closed-loop system that continuously refines itself based on past behaviors, personalization should integrate opportunities for users to encounter the unfamiliar.
Much like my aunt handing me a CD of a band I’d never heard of, platforms should introduce a degree of randomness — occasional recommendations that fall outside of a user’s habitual selections.
This could mean offering an “Exploration Mode”, where users can temporarily suspend hyper-personalization and engage with a broader range of content.
Some popular platforms do offer ways to explore beyond personalization, but none fully suspend it. YouTube, Spotify, Twitter/X, Reddit, and Google have features that broaden content, but algorithms still heavily shape recommendations.
We could push beyond an “Exploration Mode” with bold, disruptive ideas that encourage real-world engagement. For example, a “Swap Preferences Mode” could let users temporarily adopt someone else’s content profile. You might discover that buttoned-up Steve from Accounts Receivable has a deep love for death metal — an instant conversation starter.
Or take it a step further with “Blackout Mode” — a designated period when certain digital services are deliberately unavailable, nudging users to engage with the offline world.
Imagine once again flipping through radio stations or TV channels, stumbling upon something new and unexpected. Or picking up the phone to reconnect with an old friend, leading to a meaningful conversation. Even browsing a random book or magazine could spark new interests. This kind of friction adds variety to our lives — something algorithms struggle to replicate.
Of course, these features would require an opt-in preference — but the key point is that we need to explore creative ways to ensure exposure to new content beyond our comfort zones.
Balancing Convenience and Growth
Personalization in UX can unquestionably boost satisfaction and efficiency, but unchecked optimization can also confine users within self-reinforcing bubbles.
AI-driven platforms that rely heavily on past preferences risk stifling the ability to discover, grow, and adapt. In such a world of hyper-personalization, you can only enjoy your “Cake” if you already know exactly what it tastes like — leaving little room for new flavors.
For UX designers, the real challenge is finding balance. By leveraging AI without sacrificing spontaneity, we can preserve the magic of discovery. If UX is to remain truly user-centric, it must not only serve who users are today but also create opportunities for who they might become.
Don’t miss out! Join my email list and receive the latest content.
The paradox of UX personalization was originally published in UX Collective on Medium, where people are continuing the conversation by highlighting and responding to this story.
