Dive Club 🤿 – Episode Summary

Episode: Emily Campbell – AI UX Deep Dive

Host: Ridd
Guest: Emily Campbell (VP of Design, HackerRank, Creator of Shape of AI)
Date: November 24, 2025

Overview

In this episode, Ridd sits down with Emily Campbell, a leading thinker in AI user experience and the creator of Shape of AI, an influential database of AI design patterns. The conversation is a deep dive into how designers can build great AI experiences, how design patterns are evolving in the age of AI, and what the future holds for UX practice. Emily elaborates on approaches, frameworks, emerging patterns, and the ethical challenges that come with designing for human-AI interaction.

Key Discussion Points & Insights

1. Paradigm Shift: From "Designer in the Loop" to "Human in the Loop"

• AI changes the software paradigm:
  Traditional software was about designers guessing what users want, then iterating. With AI, users interact directly with the system, communicating intent and checking if the model understands.
  • "It's like we've almost shifted from designers in the loop to now this human in the loop model." – Emily (01:50)
• Role of UX designer:
  No longer just creating a fixed interface, but facilitating a "meeting place" between user and AI. The designer's job becomes helping users communicate effectively with the model, building context and trust.

2. Cataloguing AI UX Patterns

• Emergence of Patterns:
  Emily discusses her systematic collection of AI interaction patterns into categories that help teams build common language and design approaches.
  • Wayfinders: Help users onboarding with AI—e.g., sample galleries, prompt galleries, showing other users' successes to inspire first steps. (03:10)
  • Tuners: Let users refine, enhance, or clarify their prompt, often with AI-generated suggestions or confirmation steps. (04:19)
• Continuous onboarding:
  AI products inherently keep onboarding users as the system learns and new features/context become relevant.
• Trust-building through iteration:
  Direct, repeated interaction builds trust as users see the AI "getting" their intent.

3. Analogies for Human-AI Collaboration

• Hiring a human vs. using an AI:
  Emily draws an analogy to onboarding a personal assistant, demonstrating the value of verifying and tuning AI output, not just "hoping for the best."
  • "I wouldn't expect [an assistant] to go out and email my accountant or my best friend in my voice before I had a chance to see their work." – Emily (13:20)
• Skeuomorphism in AI UX:
  Early AI interfaces mirror traditional metaphors (forms, buttons), but as trust grows, UI fades and users interact with context and nuance, much like real relationships.

4. Evolving Deliverables & Designer Ownership

• As control shifts to users (and sometimes to AI), it's less clear what "deliverables" or even "ownership" design teams maintain.
  • "What does that deliverable even look like? What are we creating? And I don't know. I have more questions than answers still at that level." – Ridd (09:25)

5. Emergent AI UX Trends & Patterns

• Control for non-technical users:
  Emily highlights tools and patterns (e.g., prompt enhancers, parameter selectors, model choosers) that empower non-technical audiences.
  • "Anything that gives humans control and particularly gives humans control who aren’t super technical... that's the most interesting thing to me." (18:42)
• Parameter selectors:
  Sliders for "temperature," style, variety—giving users agency over how creative/adventurous the model gets (21:50)
• Model selection guidance:
  Clear descriptions and recommendations to demystify technical options (23:30)

6. Trust and Transparency in AI Experiences

• Building trust through transparency:
  • "Stream of thought" UIs: Making the AI's process visible; e.g., showing plan-of-action before generating output. Users must understand and trust the logic behind model decisions.
  • "The new usability becomes: how quickly can you build trust in a legible way so the user knows, 'okay, something’s happening that I can understand.'" (24:43)
• Consent and privacy as design challenges:
  Few products handle consent well; many offload risk to users instead of providing real agency.
  • Example: Limitless pendant feature only recording with verbal consent—a positive but rare pattern (29:40)
• Broader impact of data collection:
  Designers must consider ripple effects beyond the immediate user, given the depth of personal/contextual information these systems collect.

7. Evolving Design Leadership & Process

• Service-first, then software:
  Emily’s team starts by defining what a great experience with a human would look like, then adapts those patterns to AI and software layers.
  • "So we've been trying to get designs into code as fast as possible... because the model itself is now part of the experience." (33:00)
• Closer designer-engineer collaboration:
  Living prototypes and tight UX-dev feedback loops. Increasing use of prototyping tools and goal to have all designers work directly in code or code-adjacent environments by 2026. (34:32)

8. Recruiting and Skills for Modern AI Design Teams

• Curiosity is the #1 skill:
  • "The most important skill is curiosity right now. Curiosity, and then, very quickly, a go-get-it attitude." (37:06)
• Taste-building:
  Translating aesthetic sensibility into practical frameworks and opinions about what works and why, via hands-on experimentation.
• Brand expands into AI agent personality:
  Brand designers now shape not just websites, but the personality and trust layer of AI products.
• Comfort with ambiguity:
  Essential for designers and leaders; ability to facilitate open-ended exploration and collaborate across uncertainty.

Notable Quotes & Moments

• "It's like we've almost shifted from designers in the loop to now this human in the loop model."
  Emily, 01:50 – On fundamental shifts in how product experience is shaped.

• "What we're doing is not just building software for humans, we're building a meeting place between a human and something synthetic."
  Emily, 05:00 – On the profound change AI brings to UX.

• "I can just do it"
  Emily, 08:51 – On empowering users to shape experiences without waiting for designers/PMs.

• "Having great taste isn't just knowing that food is good, it's knowing whether it needs a little more salt."
  Emily, 10:52 – On building practical, nuanced design sense.

• "It's such a simple mental model... as a designer, you are creating a meeting space, facilitating this interaction."
  Ridd, 15:09 – Appreciating Emily's clarity about the designer's new role.

• "It's also so dark and bad and scary. And it just points to the importance of us as designers just to kind of put a bow on it... knowing what's below the surface."
  Emily, 30:38 – On ethical responsibility in data-rich AI experiences.

• "The most important skill is curiosity right now... followed very quickly by go get an attitude."
  Emily, 37:08 – On emerging skills for hiring in AI UX.

Notable Timestamps for Key Segments

• 01:50 – Emily explains the "human in the loop" shift.
• 03:10 – Introduction to "Wayfinders" in AI onboarding patterns.
• 04:19 – "Tuners" and letting users clarify AI prompts.
• 13:20 – Onboarding AI as you would a personal assistant.
• 21:50 – Parameter selectors, temperature sliders, control in MidJourney.
• 24:43 – Trust, transparency, "stream of thought" UI, and consent.
• 29:40 – Data/consent challenges; Limitless pendant example.
• 31:42 – Designing experiences: service-first, software-second.
• 34:32 – Moving faster into code, living prototypes, evolving collaboration.
• 37:06 – What Emily looks for in new design hires: curiosity and taste.

Tone & Style

The episode maintains an inquisitive, reflective, and practical tone. Emily balances optimism about UX innovation with grounded caution regarding ethical and privacy challenges. The language stays approachable yet insightful—rich with analogies, metaphors, and lived experience.

Summary

This Dive Club episode with Emily Campbell offers a blueprint for thinking about AI UX—not just as interface tweaks, but as a new paradigm that centers user intent, trust, and agency. Listeners are left with frameworks, practical examples, and challenging questions about consent, privacy, and the expanded scope of design. Emily’s advice for designers ("just go build," "lean into curiosity," "invite others into ambiguity") is refreshingly candid and actionable.

For more key takeaways and future episodes, visit Dive.club.