AI UX DAILY
Tuesday, June 30, 2026
4 stories · curated for designers
The stories
Today in AI Products
 Emily Campbell / TLDR Design |
Jun 29 |
A six-layer framework for designing AI experiences below the interface
Designer and researcher Emily Campbell argues that generative AI has made product systems probabilistic, and the old deterministic design model no longer holds. She proposes six interdependent layers, including context, harness, model, governance, and emergence, that sit beneath the visible UI. The point is that design decisions now live across all of them, not just the surface.
| “ |
Map your current AI product against these six layers and identify which ones your team has no design input on yet, because those are likely where your worst user surprises are coming from. — Designer's Takeaway |
| Vispero / TLDR Design |
Jun 29 |
WCAG compliance and real accessibility are not the same thing
A new piece from Vispero lays out a concrete case that a product can pass every WCAG checkpoint and still be effectively unusable for people with disabilities. The wheelchair ramp blocked by a telephone pole analogy is blunt and accurate: technical conformance does not equal usable access. The article calls out the gap between checkbox audits and genuine usability testing with disabled users.
| “ |
Schedule one usability session with a screen reader or switch-access user on your current AI-powered flow before your next release, because a passing audit will not catch the problems they will find in the first five minutes. — Designer's Takeaway |
| MIT Technology Review |
Jun 29 |
Naming your AI agent 'Alex' is a design choice with real consequences
MIT Technology Review examines the growing trend of companies giving AI agents human names, pronouns, and coworker framing. The piece argues this is not neutral branding: it shapes user expectations about reliability, accountability, and what happens when the agent fails. When something goes wrong, users are left confused about who or what is responsible.
| “ |
Audit the name, persona, and role language you are using for any AI agent in your product and ask whether the framing accurately predicts the agent's actual capabilities and failure modes, because mismatched expectations are a trust problem you will have to design around later. — Designer's Takeaway |
 Perplexity |
Jun 28 |
Perplexity launches a legal AI platform built on a 20-model agent stack
Perplexity has released a legal-focused AI platform aimed at law firm workflows, positioning it as a competitor to Westlaw. The system uses a 20-model agent architecture to handle research, document review, and related tasks. It is a concrete example of a general-purpose AI company shipping a domain-specific, high-stakes professional tool.
| “ |
Notice how Perplexity is adapting a general search and reasoning product for a domain where errors carry serious consequences, and use this as a prompt to audit where your own AI product surfaces confident-sounding answers in contexts where a wrong answer actually matters to users. — Designer's Takeaway |
Today's Idea
The interface is the smallest part of the design problem
Three of today's stories point at the same underlying truth: the visible UI is just the outermost layer of an AI product. What you name the agent, which accessibility gaps the audit misses, and which model layers your team has no input on all shape the user experience just as much as the screen does. Designers who only work on the surface are handing off the most consequential decisions to people who are not thinking about users at all.
Stop shipping AI slop
Audit your AI design against 36 patterns
Drop a screenshot, get specific gaps and a Claude Code prompt to fix them. Free, no signup for the first audit.
Audit your design →