11 Comments
User's avatar
Ryan's avatar

Love your thoughts here, Julie! We are moving towards reciprocity over control when it comes to multi-modal conversational interfaces which ultimately leads to designing relationships with reflexive intelligence or intelligent systems.

Expand full comment
Subhash's avatar

lego is my spirit toy too

Expand full comment
David Delaney's avatar

Great read. Imagine if every child could have an education tailored exactly to how they understand and learn. What would that generation look like.

Expand full comment
Gortin Shyver's avatar

There’s nothing new here and it’s so bloody long all your points were lost in the ocean of examples. Who gives a whit about Whitman’s worn-out quote on multitudes or a spit about Stewart’s opinion on obscenity? If you want to write a novel, write a novel.

Expand full comment
Samuel Payne's avatar

Loved this, Julie. Made me think, so I wrote it down. I hope you don't mind me quoting you: https://www.readwireframe.com/p/sending-a-fax-from-the-beach

Expand full comment
Reinvent Critically's avatar

Thanks for writing this piece. It’s full of sharp insights. I’ve been exploring similar questions lately, especially around how AI can better respond to human input through more adaptive interfaces.

I’ve been coming at it from the semantic side. I believe that before we can design flexible outputs, we need a structure that helps systems understand what the user is trying to do. That’s what led me to write a three-part series on building a semantic layer that can detect intent, label tasks, and generate UI based on how the conversation shifts.

After reading your essay, I felt like we’re circling similar ideas from different angles. I’d love to share the first part of what I’ve been working on. If it’s interesting to you, I’d really appreciate any thoughts.

https://jinn1.substack.com/p/designing-the-semantic-layer-for

Expand full comment
Callum Mcleod's avatar

Great post! Little typo when you referred to WYSIWYG, you missed the last Y 😃

Expand full comment
David Mendes's avatar

When using tools like ChatGPT, I often feel like pulling my hair (if I had any) because how cumbersome it feels when steering complex tasks. Maybe its the right moment for GenerativeUI. Options and controls that materialize themselves during the conversation, somehow predicting what kind of additional controls the user might need.

Expand full comment
Samantha's avatar

So true especially when you’re looking for a visual output. I just want to click in it and move things around, and I can’t. I have to go back to words. It’s like trying to tell someone where to scratch an itch on your back. It’s so much easier to point.

Expand full comment
Ami Vora's avatar

Another great post, Julie. I'm loving this AI series!! It's so good!

Expand full comment
Julie Zhuo's avatar

Means a lot to me to hear you say that Ami!

Expand full comment