According to Fast Company, Figma is developing its AI strategy by closely observing its largest customers, many of which are technology companies building their own AI integrations. The company has identified two core industry trends it is pursuing: augmenting existing user workflows with AI to save time and add capabilities, and creating entirely new experiences where AI is the central component. Interestingly, Figma’s market research suggests users actually value AI more when it’s woven into familiar workflows, even though it becomes less visible. The core design challenge, as framed by Figma, is akin to the frictionless experience of a Google homepage, where even a single pixel of unnecessary friction can impact millions of users daily. The fundamental goal is to introduce these powerful AI features without getting in people’s way or disrupting their creative flow.
The Invisible Assistant
Here’s the thing about Figma’s first approach—adding AI to existing workflows—it’s kind of brilliant in its subtlety. You’re not being asked to learn a whole new tool or change how you work. Instead, the AI just makes the things you already do faster or unlocks a next step you couldn’t easily do before. Think auto-complete for design, or a one-click tool that generates a set of icons based on your text prompt. The value is huge, but the AI itself fades into the background. It becomes a utility, like electricity. You don’t think about the power grid when you flip a light switch, you just get light. That’s the ideal state for this kind of integration: powerful but invisible.
The New Frontier
But then there’s the second path: building totally new AI-native experiences. This is where things get speculative and, frankly, more risky. We’re not talking about a better button here. We’re talking about potentially reimagining the entire design process from the ground up with an AI co-pilot. What does that even look like? A conversational interface where you describe a user flow and it drafts the screens? A system that can ingest product requirements and spit out a prototype? This is uncharted territory, and it’s where Figma could either leapfrog the competition or build something nobody actually uses. The trick will be making these new experiences feel as intuitive and frictionless as the canvas people already love.
Learning from the Builders
What’s really smart is how Figma is learning. Their biggest customers aren’t just banks or retailers; they’re tech companies building AI into their own products. That’s a massive feedback loop. Figma gets a front-row seat to what’s working and what’s failing in real-world, large-scale AI implementation. That kind of insight is pure gold. It probably means their AI features will be less about flashy demos and more about solving genuine, painful bottlenecks their most demanding users face every day. It’s a built-in reality check.
The Friction Paradox
Now, the most fascinating challenge they mentioned is the friction paradox. Their canvas is a high-stakes environment. Introduce a clumsy new panel or a confusing prompt box, and you’ve just wasted the collective time of a massive, global user base. So how do you introduce something as fundamentally disruptive as AI without causing disruption? It’s a tough needle to thread. You have to educate users about new capabilities without annoying them. You have to make the AI helpful but not intrusive. It seems like their research gives them a clue: bake it into the workflow so seamlessly that the workflow itself feels upgraded, not interrupted. Easier said than done, but if anyone has the design chops to pull it off, it’s probably Figma.
