Best Practices
Case studies and articles on AI in design — real workflows from leading studios and teams
Emily Campbell: The layers of AI experience
Emily Campbell proposes a six-layer model for AI product design, arguing that effective design now requires understanding the systems beneath the interface, not just the interface itself.
Smashing Magazine: How to make your design system AI-ready
Vitaly Friedman explains how to prepare a design system for AI-generated prototyping through spec files, token layers, and automated audits—with examples from Atlassian, IBM Carbon, and Nordhealth.
UX Collective: Designing for AI means designing like it's 1999
Patrick Neeman argues that AI interface design mirrors the early web era: enormous capability, almost no shared conventions, and tools that evolve faster than any standard can settle.
UX Collective: The prompt is not an interface
Joshua Leigh argues that AI prompting is a regression in interaction design — and what it means for designers and the tools they build.
Designer Fund: AI in Design 2026 — the inflection point is here
A survey of 906 designers across 60+ countries traces how the profession has rebuilt around AI, from daily tool adoption to hybrid design-engineering roles.
Figma: the design agent is here
Figma's account of its native AI agent built into the canvas — what it automates, how it reads design systems, and who gets access in the beta rollout.
Smashing Magazine: Five patterns for showing what AI agents are doing
A pattern library for agentic AI interfaces, from subtle background status updates to persistent audit trails suited to high-stakes decisions.
Microsoft Design: building an AI-forward design system for Copilot
Jon Friedman describes the Copilot Design System's four surfaces and the 'throw and catch' pattern that keeps AI context continuous across Microsoft products.
NN/g: Designing AI agents — 4 lessons from China's Qwen
Nielsen Norman Group studied the Qwen AI agent to extract four practical UX design lessons for teams building agentic experiences for new users.
Medium: UX design in 2026 — from producing to deciding
A Lead UX Designer argues that AI has shifted professional value from visual execution to strategic judgment, and that designers who ignore this shift will lose ground.
UX Collective: how to make Claude Code follow your design system in Figma
Sen Lin's May 2026 guide builds four Claude Code skills that enforce design system rules at every stage of an AI-assisted Figma session — from token preflight to QA.
Smashing Magazine: designing stable interfaces for streaming content
A practical guide to preventing layout shifts, scroll overrides, and DOM overload in AI chat and streaming data interfaces, with code-level solutions.
Medium: essential UI/UX design skills for the AI era in 2026
Dolly Borade Solanki maps three skill categories that define career resilience for UI/UX designers in 2026 — human-centric, AI-augmented, and business — as AI automates routine execution tasks.
Smashing Magazine: When AI pressure turns designers into reluctant engineers
Carrie Webster examines what happens when companies expect production-ready code as a design deliverable, and why the economics argue against it.
UX Collective: Product design in 2026 — the three walls that fell
Kike Peña argues that AI and vibe coding have removed three structural barriers that once limited designers' influence, and explains what designers must do differently now.
Nielsen Norman Group: AI agents as users
NNGroup argues that AI agents now function as users of digital interfaces, requiring designers to rethink who—and what—they are designing for.
Figma: how the FigCade was built using Make, Weave, and MCP
A Figma case study showing how Make, Weave, and the MCP server worked together to produce six browser games in a single weekend.
Smashing Magazine: When to show users what an AI agent is doing
A decision-making framework for identifying the moments when AI agents must reveal their activity, with an audit process and pattern selection rubric.
Design Systems Collective: from prompt to production with Claude Design and Claude Code
A step-by-step account of going from visual exploration in Claude Design to production-ready code via Claude Code, with design system tokens as shared context.
Smashing Magazine: A practical guide to establishing design principles
Vitaly Friedman outlines an eight-step workshop method for creating design principles that guide team decisions rather than decorating a wiki page.
UX Collective: Your AI agent knows the codebase but not the product
Gregory Muryn-Mukha on why AI coding agents need explicit product context before they work, and what that means for design and product teams.
UX Collective: becoming an AI-native designer
Sen Lin describes how Claude Code and Figma Make rebuilt his design workflow from scratch, and why the hard part is transmitting tacit knowledge to AI rather than using the tools themselves.
UX Collective: AI, UX, and the factory model
Zeeshan Khalid maps three generations of AI design integration and argues the strongest 2026 results come from hybrid human-AI workflows.
UX Collective: Agentic AI, design systems & Figma: a practical guide
Christine Vallaure examines how agentic AI shifts designer focus from full-page mockups to building blocks, using a live Storybook demo as the entry point.
Teads Engineering: how AI let designers write their own components
How connecting Figma to an AI IDE via MCP let Teads designers write component code directly, cutting multi-week backlog delays to days.
Nielsen Norman Group: GenUI vs. vibe coding — who's designing?
NNGroup's Kate Moran draws a precise line between generative UI and vibe coding, showing how each shifts design responsibility to a different actor and fails in a different way.
8 AI tools that work for designers in 2026 — practical guide
A designer's firsthand review of eight AI tools tested in real client projects, reporting which ones earned a permanent place in daily workflows.
RGD: AI tools for designers in 2026 — responsible use and creative support
The Registered Graphic Designers association publishes March 2026 guidance on selecting AI tools responsibly, with an updated Code of Ethics covering transparency, documentation, and human review.
Ultimate design workflow in 2026: from conversation to deployed product — Medium
A March 2026 Medium article documents a 7-step AI-assisted workflow where designers move from AI-guided planning through to real product deployment without writing traditional code.
John Maeda: Design in Tech Report 2026 — from UX to AX
Maeda's annual report argues that design's central challenge has shifted from helping users execute tasks to helping them evaluate whether AI agents did those tasks well.
Human strategy in AI-accelerated design workflow — Smashing Magazine analysis
An article on maintaining strategic thinking and human judgment as AI takes over more execution tasks in design workflows.
Parallel HQ: How to integrate AI into your creative design process
A practitioner's guide to five AI applications in design workflows, with case studies from IBM, Studio Rx, and Paramount+, and a framework for measuring actual impact.
NN/G: the design process is compressed, not dead
Nielsen Norman Group argues that experienced designers compress design process rather than skip it, and that AI tools accelerate this compression without replacing the judgment behind it.
Figma: 5 design skills to sharpen in the AI era
Figma's research-backed look at which capabilities matter most as AI changes what product teams can accomplish without specialized design training.
AI image tooling and interactive prototyping — Figma workflow case study
Figma's own workflow guide showing how to combine AI image generation with interactive prototyping for richer design exploration and testing.
Design is how it wins — UX Collective analysis
An essay arguing that in an AI era where functional UI is commoditized, the designer's value shifts from making things work to making them win in the market.
Medium: the AI-augmented designer — what changed in 2026
Raj Paul describes how the designer-to-developer workflow has collapsed from weeks to hours using MagicPath and Claude Code, and what it takes to ship a functional prototype from a Figma file in under an hour.
State of the Designer 2026 — Figma survey analysis
Figma's annual report on how designers are adapting to AI, based on survey data from thousands of practitioners navigating the transition.
Medium: What Punch taught me about AI design workflows
Pranit Jena tests six AI tools on a single game project to map which tools think and which ones build, then draws practical conclusions for design workflow choices.
Activated Thinker: Google Stitch + Antigravity — the design-first AI workflow for 2026
A hands-on walkthrough of using Google Stitch and Antigravity together to audit and build landing pages starting from design intent rather than code.
Jeff Zych: Claude Code is reshaping my design process
A product designer describes replacing wireframing with AI-generated prototypes in code, noting faster UX feedback and fewer Figma sessions for everyday feature work.
Integrating AI into design system workflows — Smashing Magazine analysis
A practical guide on making design systems AI-ready, covering token structure, component documentation, and MCP integration for AI-assisted design.
AI changing design hiring and portfolios — UX Collective analysis
An analysis of how AI tool proficiency is reshaping design job requirements and what designers should demonstrate in portfolios to stay competitive.
Medium: How I'm actually using AI in my UI design workflow in 2026
A UI designer documents two practical AI applications—concept generation and QA automation—and explains what AI still cannot do in production design.
How design leads use AI — UX Collective case study
An analysis of how design leads are delegating the 80% of work that is execution to AI, reserving human effort for the 20% that requires judgment and strategy.
Smashing Magazine: Designing for agentic AI — practical UX patterns for control, consent, and accountability
Victor Yocco outlines six UX patterns — intent preview, autonomy dial, explainable rationale, and more — for building agentic AI systems that users can trust.
Muzli: 15 AI design tools that actually change how you work in 2026
A curated tool list tested in production contexts and organized by workflow stage, with recommendations segmented by designer type — solo practitioners, team leads, and designer-developers.
Medium: My AI design workflow — what actually works in 2026
A senior product designer's account of testing 40+ AI tools and isolating the small subset that survived contact with a real workflow.
Medium: AI in product design — where we are now in 2026
David Robinson's reality check on AI in product design: what changed, what didn't, and why design systems are now built for machines.
Vibe coding as a UX design workflow — practical guide
A designer's account of incorporating vibe coding into UX work, covering what works, what fails, and where the technique fits in professional design practice.
UX Collective: Are we doing UX for AI the right way?
Katya Korovkina argues that chatbot-first thinking is the dominant misconception in AI UX design, and explains when conversational interfaces fail.
UX Collective: The most popular experience design trends of 2026
Joe Smiley surveys nine experience design directions gaining traction in 2026, from multimodal interfaces and machine-readable design systems to the return of glassmorphism.
Medium: Designing for 2026 — process, agency, and decision quality in the age of AI
VP of Design at OneSignal argues that AI has compressed timelines without reducing the need for rigorous thinking, and explains why craft has become the main differentiator.
UX Collective: how top companies are using AI in their design workflows
Punit Chawla examines how enterprises like Atlassian and Meta have restructured their design processes around AI at the organizational level.
Smashing Magazine: Beyond generative — the rise of agentic AI and user-centric design
Victor Yocco argues that designing agentic AI systems requires an entirely new research playbook, shifting UX from usability toward trust, consent, and accountability.
AI in UX design and research — NN/g case study
NNGroup's evidence-based guidance on where AI adds value in UX research and design processes, based on their ongoing studies of design team practices.
NN/g: Humanizing AI is a trap
Nielsen Norman Group argues that designing AI systems to appear human-like backfires: it inflates user expectations, reduces utility, and creates relational risks.
House of gAi: AI for graphic designers in 2026 — how to integrate AI into a real design workflow
A practical Medium essay arguing AI has moved from novelty to workflow component, with guidance on where to place it and what to keep human.
UX Collective: 10 UX design shifts you can't ignore in 2026
Arin Bhowmick, Chief Design Officer at SAP, identifies ten structural shifts in UX practice shaped by AI integration, agentic systems, and evolving user expectations.
UX Collective: Why AI is exposing design's craft crisis
Dolphia argues that AI didn't create the design industry's technical literacy problem — it simply made visible a gap that has been widening for over a decade.
Medium: Building agentic design systems
Luis Ouriach argues that design systems need to expand from component libraries into AI-enhanced product systems spanning all brand touchpoints.
Figma: five shifts redefining design systems in the AI era
Figma identifies five structural changes in how design systems work as AI becomes routine — from encoding craft to governing agent workflows.
UX Collective: structuring your design system so AI can read it
Oleksandra Huba explains why screenshot prompts fail with LLMs and what Figma naming conventions, Code Connect, and three rules files actually fix.