Welcome back, designer. This edition dives into the intersection of usability and user rights like privacy. Plus, a relevant book rec, design career tip, and a nod to the iPhone notch.
Note: this issue has sponsors including Interaction Design Foundation.
[USABILITY CRIT]
Stop hiding my controls
Read this in-depth teardown of how “conceal-and-reveal” types of user interfaces hurt usability, especially for older users.
Here’s an example of of a UI that hides its controls:
To keep picking on Apple, here’s a quote from the article:
If I want to activate the flashlight on my iPhone, I have to know to swipe up from the bottom left-hand corner in order to bring up a control panel where the flashlight button exists.
There is absolutely nothing on the interface that indicates this is an available action.
…It is the quintessential hidden control—useful, lurking, and available—but only to those who know the trick.
Some other takeaways:
Lab tests found this type of UI adds 1.7s to average to locate hidden settings
Contrast drops when buttons float over dynamic content
Recommends progressive disclosure adjacent to focus, not nested menus.
Try this: audit your product’s top 3 flows. Count hidden versus visible controls, and aim for less than 20% hidden. Medical and enterprise/internal level UIs are full of these opportunities.
This bit of usability dovetails nicely into the next big story…
[AI ETHICS]
Meta AI’s accidentally shares private chats;reveals its terrible privacy design
Meta AI has gone viral recently, and for a bad reason: the company shared private chats for the public to discover.
Worried about my own data leaking, I searched to see how I could turn this off.
In an eerily similar feeling to Stop hiding my controls, I had to count 5 taps to apply this update:
Log into Meta AI, go to settings gear icon
Tap on the first option, “Settings” which is not that obvious
Tap on “Data & Privacy”
Scroll to the bottom, then tap “Make all prompts only visible only to you”
Then, finally, tap “Apply to all”
The thing is that when I first signed up for Meta AI - logging in with my Facebook account - I didn’t see this presented as an option, at all.
So of course thousands of participants unknowingly ended up sharing their prompts and private chats.
As if that’s not enough, Meta may be coming for your unpublished photos (Verge).
Meta is testing a feature that uploads your entire camera roll so its models can auto-create collages and “moments.”
This upcoming feature is supposed to be opt-in, but some users have already found that Meta has violated photo privacy.
In this Reddit thread, someone’s wedding photos go Ghibli-fied without their consent:
“This photo from a post I made years ago popped up in my stories...in AI form. There's no way in the settings (that I've found) to stop it from doing this.”
Takeaway: If you’re designing a new experience involving people’s personal data, 100% consider an “opt in first” approach in the user experience. Because that “opt out first” dark pattern may end up in billions of dollars in lawsuits…
All of this leaves me even more perplexed, because Meta is getting itself into a hot mess, for features that I don’t think should be its strengths
Meta AI is huge missed opportunity
OK OK…one last thing on Meta AI. When it first made a buzz with its launch, this is the differentiated use case that I had hoped Meta AI would provide:
Instead, Meta AI is just another competitor to chat LLMs like ChatGPT, Gemini, Claude.
Imagine if Meta AI actually leaned into its power it spent building over the last 2 decades…it’s social graph.
I don’t want to create images with filters. That’s so 2024.
I want Meta AI to proactively curate social opportunities for me:
Help me manage my social calendar
Surface up relevant events and Facebook groups
Suggest friends who are in town or who might have moved
Help me engage with people who have liked my posts the most
I will gladly opt into a beta if Meta just asked me: “Hey, would you like to turn on AI superpowers to make your social life easier?”
I think this is a HUGE miss. What do you think?
SPONSORFor those looking for continuing UX education, or something lighter-weight than a bootcamp, then check out Interaction Design Foundation.
Their new course, AI for UX Designers, is a huge level up for modern designers.
Use my partner link for a 7 day free trial.
[DESIGN CAREER CORNER]
Try the “spelunking” method
Ex-23andMe design leader
swears by this deceptively simple hack: go on a deep dive of team drives, legacy folders, and old docs + analytics.
Just as “spelunking” is the exploration of caves, spelunking through your org’s files can unearth past tests, customer rants, and half-built ideas that all provide great context, especially for new teams and projects.
Other benefits of this approach:
Speeds ramp up time, avoids rediscovering known failures.
Turns you into the context hero in week one.
Doubles as low-lift portfolio fodder (“here’s what we didn’t build, and why”).
Imagine if some proper spelunking could reveal past product decisions that have led to massive ethics issues and lawsuits…
Try this: schedule a 2-hour “archive dive” on your next project. Then share one forgotten insight with the team.
[DESIGN BOOK REC]
The Goods of Design: Professional Ethics for Designers
The Meta AI story has me thinking about design ethics, again. The author argues that design is a practice (like law or medicine) with its own “internal goods.”
These goods (human flourishing, environmental stewardship, honest communication, equitable access) give the profession a moral direction that goes beyond client briefs or market KPIs.
This book sets out to answer three deceptively simple questions:
What ends should designers pursue?
Why do those ends matter for society and the planet?
How can designers develop a shared ethical compass without a formal Hippocratic-style oath?
Why it matters for today’s designers:
Design decisions scale globally. From dark-pattern checkout flows to privacy-encroaching AI models. Without a shared ethos, “move fast and break things” remains the default.
Existing guidelines are patchy. Unlike nursing or architecture, design lacks an enforceable professional charter; Guersenzvaig offers a starting blueprint.
Virtue over compliance. Instead of a checklist, the book nudges practitioners to become the kind of people who habitually do the right thing—even when no one is watching.
[DESIGN HISTORY TIDBIT]
The iPhone notch is 7 (almost 8) years old
Sept 2017: iPhone X debuts the “sensor housing.”
Apple kills the Home button and carves a bite-shaped peninsula out of the screen, triggering instant memes and tweets.
The impact was huge:
With no Home button, fluid edge-swipes become the primary nav model; designers scramble to drop bottom toolbars and rethink reachability. (mjtsai.com)
The notch paved the path for 2022’s Dynamic Island and every selfie-sensor hole-punch since. Proof that odd hardware bumps can evolve into new UI real estate.
Takeaway: planning layouts for Vision Pro, foldables or the next new device? The notch lesson holds: design with hardware quirks, not against them.
Which nugget stood out most to you?
If you enjoyed this article, help spread the word by leaving a comment, sharing the post, or even DM’ing me what your favorite user experience is. I love hearing from you.
Stay curious, designer!
Written by
, also author behind Money for Humans
Oz, thanks for including my post and sharing the "spelunking" tip with your followers. 🙌🏼