Figma’s guide to spending fewer AI credits with the agent: scope prompts clearly, attach only the context the task needs, and start a new chat when you switch tasks, since long chats raise the cost of every prompt. For repeated jobs, turn the work into a generative plugin — building it costs credits, but every run after that costs zero.

Enterprise admins can now set default per-user access to paid AI credits at the organization and billing group levels, and manage credit access and limits through SCIM. They can also block people from creating Starter or Professional plans with company emails, and use new admin REST APIs to manage teams, folders, and files. Enterprise plans only, rolling out in the coming days.

On the All-In podcast, Dylan Field takes back his own Config prediction: ​“I said the ceiling will be raised by AI… That was wrong.” AI lowers the floor for making code, motion, and design, but the ceiling only rises when designers ​“choose to focus, invest their time, and to be bold.” With exponentially more software and flat app usage, that is now where products differentiate.

Figma’s newest product is a Markdown editor. ​“You can think of Specs as a container for your context.” Specs live on the canvas and as their own file type, are multiplayer with comments, and can be filled by hand or through MCP and GitHub. Rolling out in open beta on paid plans.

“Intelligent UI is a step toward a ChatGPT that can shape an interface around what you’re trying to do. Over time, we want ChatGPT to create more of the interfaces people need for everyday tasks. You should be able to describe what you want to accomplish and work with an experience built around it.” OpenAI’s take on generative UI ships to all ChatGPT users, built on a component library plus a model trained to judge layout and interaction.

Gemini 4 Argon, Google’s latest model, is now in the Figma Make model picker on all plans.

A checklist for spotting AI-generated interfaces, written after a college app’s ​“minor UI improvements” update. Gradients everywhere, pulsing ​“active” badges that can never be inactive, fingernail cards, glassmorphism, and taglines built from ​“Elevate”, ​“Seamless” and ​“Unleash”.

Jeff Smith of Fieldwork benchmarked frontier and open-weight models on adding icons to an existing set, with humans grading recognizability, process, consistency and execution. His summary: ​“they’re steadily improving execution, their choices often overly literal.”

Evil Martians’ eight-part framework for keeping AI-built interfaces consistent. Component contracts say when to use and not use each component, an inventory lets agents shortlist components before reading any code, and a design linter fails the build on a raw hex color. The agent skill is published too, split into craft and use workflows.

Anton Lovchikov, Head of Design at Evil Martians, makes the case that ​“a designer’s main job isn’t drawing screens. It’s pulling rules out of chaos and making sure they’re followed.” His proof is a 700-line localization refactor, after which Claude rewrote half the strings. The fix was a three-layer system of product principles, copy rules traced back to them, and skills built on those rules. Now when he dislikes a string, he asks Claude which rule produced it and fixes the rule.

Why stay in the details when agents can do the hard part? Geoffrey Litt, design engineer at Notion, says the time spent in the nooks and crannies of a problem loads your memory with the raw material your brain recombines later, even in the background. His choice for the AI era: get lazy and settle for good-enough, or work harder than ever. ​“You can already start to tell who is taking each path.”

Brett McMillin, Designer Advocate at Figma, runs a skill that pulls user personas from a Notion page and critiques selected frames as one persona, in his demo a fintech app for teens. The findings land as annotations on the canvas covering comprehension, findability, visual hierarchy and trust, and the skill is told never to touch the design. ​“Nothing will take the place of actual user testing and talking to our customers.”

Peng Zheng’s personal site updates itself, as he shows host Claire Vo on How I AI. Send his check-in bot a photo, a screenshot or just the name of a place, and it looks the place up through the Google Places API, fixes the perspective, removes people and generates light and dark 3D miniature versions for the site.

Main-stage talks from Lenny & Friends Summit, with Marty Cagan on strong opinions loosely held and Claire Vo on whether any of the extra shipping matters. Start with Katie Dill of Stripe, who warns that AI building could leave us with ​“zombie UI that’s monotonous, vacant, or uncared for.”

In 1906, John Philip Sousa warned that the phonograph would become ​“a substitute for human skill, intelligence and soul.” Jeffrey Katzenberg traces a line from Sousa to the fight over ​“canned music” in sound film to his own decision to end hand-drawn animation at DreamWorks: ​“This has happened many times before, and it was never settled by the technology. It was settled by the terms.” His line on today’s models: ​“It is statistics, not soul; it is emulating things that have been done.”

“A guideline can describe the brand. It cannot participate in the work.” Paul Jun, who leads brand at Ramp, on why in-house brand teams turned into service desks and what should replace them. Ramp’s version already works – you @mention an agent in Slack, attach a doc, and get a finished deck back in HTML, PDF, and PowerPoint. His proposed team pairs artists who set the bar with builders who turn it into components, agents, and evals. ​“Taste remains human. Repetition becomes software.”

Will Newton, Staff Product Designer at Ramp, wrote up the demo he gave at the Design x AI event. His ​“design harness” is a folder with two text files: Rules​.md tells the agent how to behave and what to never do, Design​.md holds colors, type scale, spacing, component styles, and the philosophy behind them. Everything else (Paper, Figma via MCP, HTML prototypes, production code) hangs off that folder. No app, no plugin – just plain text an agent reads before it touches anything.

ChatGPT now has Sketch (type “@Sketch”, draw a rough layout, get it rendered), templates for formats like posters and merch, comments pinned directly onto an image for focused edits, and shareable prompts so someone else can run your idea on their own photos. That’s a design tool’s vocabulary showing up inside a chat. Generation latency is reduced by up to 50% compared with Images 2.0, it’s live on every ChatGPT tier including free, and developers get two new API models: Flare as the default, Sunburst for precision work with longer generation times.

“I worry a little bit about function following form.” That’s Katarina Batina, VP of Design at Shopify, in the middle of a piece about how completely her team reorganized around AI tooling. Roughly 60% of Shopify’s designers now push code directly to production, up 38% since last September, with over 5,500 live prototype sites built on Quick, the internal hosting platform. Her caveat is the part worth keeping: ​“There’s so much discourse around what tools are in fashion this week versus next week. The problem is we risk losing connection with the questions we’re trying to answer and the people we are building software for.”

The skills demo is the one to watch. Rodrigo calls a slash command for a skill his teammate Tammy wrote in plain English, carrying her own design-review conventions, and the agent returns the critique as annotations right on the canvas. Team practice becomes a callable command.