Log in or create a free Rosenverse account to watch this video.
Log in Create free account100s of community videos are available to free members. Conference talks are generally available to Gold members.
Summary
What are AI-mediated experiences made of? What new interactions and UI patterns should be part of your toolkit? How do these new patterns support trust, critical thinking, usability, and accessibility? Watch Josh Clark and Veronika Kindred, authors of our forthcoming book Sentient Design, explore emerging best practices for the design of machine-intelligent experiences. The session focuses on the practical: new interaction patterns, functional patterns, and UI patterns for AI-powered interfaces. Meet the Pinocchio pattern, see how trait tags work, learn to nudge, become a master of inpainting, and so much more. Learn how these solutions are tailored to suit the curious qualities of machine intelligence. You’ll see how they build healthy mental models for users, creating realistic expectations for system capabilities. Plus, learn to use these new patterns to guide behaviors in ways that amplify user judgment and agency, instead of replacing them.
Key Insights
-
•
Open-ended AI prompts create a discovery deficit, making it hard for users to know what to ask or how to start.
-
•
Nudge patterns, such as prompt chips and trait tags, help users explore AI capabilities and refine outputs collaboratively.
-
•
Version vaults help manage generation overload by saving, comparing, and reverting multiple AI outputs, encouraging experimentation.
-
•
LLMs produce probabilistic, not deterministic answers, so users and designers must accommodate uncertainty rather than expect singular truths.
-
•
Scenario patterns surface alternative AI-generated responses, helping users understand the range of plausible answers and avoid illusions of one true answer.
-
•
Adventure navigation patterns introduce user agency in algorithmic feeds, allowing users to branch and personalize content journeys.
-
•
AI-generated results can be confidently wrong, posing risks that require UI patterns to signal uncertainty and prompt human judgment.
-
•
Suggestion-first design ethos promotes offering users direction without imposing decisions, balancing control and convenience.
-
•
Citation patterns increase transparency and trust by showing sources behind AI-generated information.
-
•
Active engagement in shaping AI tools and interfaces is critical as AI tech is nascent and its social impacts uncertain.
Notable Quotes
"AI is supposed to make things easier, but it usually comes on with some knock on problems."
"We keep asking these nondeterministic systems deterministic questions."
"Nudges are subtle ways of drawing attention and giving examples of what you can ask AI."
"Version vaults act like a safety net to encourage more exploration by tracking all your generated versions."
"LLMs are always confident, even when they are wrong, making it hard to tell the difference."
"Sentient design experiences propose direction without imposing it, letting users stay in control."
"There’s rarely one true answer, especially in machine-generated results, so we should surface relative confidence instead."
"Adventure nav lets users choose their own path through content, offering freedom rather than algorithmic sameness."
"When systems detect something unusual or uncertain, that’s when human agency and critical thinking is necessary."
"If you’re fearful or skeptical of AI, that makes sense, but now is the moment to get involved and shape the future."
Or choose a question:
More Videos
"Data shifts the conversation from I think to I know, but bad data can be dangerous without the right skepticism."
Steve Portigal Chris Chapo Kelly Goto Christian RohrerDiscussion
May 13, 2015
"If partnering with design makes it easier to meet product framework requirements, then everyone wins."
Jess GrecoCreating a Basis for Change: Scaling Design Maturity
June 8, 2022
"We study failure without blame; this is the path to progress."
Dan WardFailure Friday #1 with Dan Ward
February 7, 2025
"Question marks for people who build things aren’t that great, but they’re also what we look for."
Husani OakleyBias Towards Action: Building Teams that Build Work
June 14, 2018
"The best large language models we tested only achieved about 47% accuracy in categorizing Reddit posts."
Karen McGrane Jeff EatonAI for Information Architects: Are the robots coming for our jobs?
November 21, 2024
"There is no one size fits all. Go in, do your research, find failure points, and address those directly."
Kristin SkinnerOpening Keynote: Org Design for Design Orgs
November 6, 2017
"Design systems typically lack content on when and how to use components—adding this makes design more self-service."
Alfred KahnA Seat at the Table: Making Your Team a Strategic Partner
November 29, 2023
"Sometimes the most annoying small problems are the best places to start to build momentum."
Liza Pemstein Jane DavisScaling Research Via an Ops First Model at Clever
March 27, 2023
"That’s because you’re thinking with your research hat on, not your UX leadership hat."
Nalini P. KotamrajuTwo Jobs in One: Being a “Leader who is a Researcher” and a “Researcher who is a Leader"
March 10, 2021
