Your Design Charter Already Has a Position on AI
Many designers feel stuck or anxious about this moment of change in product development because they're being asked to adapt to a new set of AI-based tools and processes before anyone has helped them see how those tools connect to their values, or to the value designers bring to an organization. The question hidden behind all this tooling discussion isn't a new one: what do we believe design is for? If your team has done the work to articulate that in a team charter, a set of principles, or a practice philosophy, then you already have a framework for evaluating AI.
My team at Everfi created our Design Charter a few years ago. It names what we're trying to do, what we're committed to, and how we think about our value to the organization. The charter frames our process as something that reduces risk, and increases organizational flexibility and innovation, not just a step in a delivery pipeline (a way of thinking that the whole org benefits from). It commits us to exploring and sharing possible visions of the future of learning. Ultimately the charter articulates our thinking about why design matters to the company.
When AI tooling started generating real noise around the industry — the excitement, the anxiety, the pressure to have an answer — I went to our charter, because it's the place where my team has already settled the question of what we're here for. It felt more honest to start there than to let the conversation get pulled toward whichever tool had the most momentum that week on LinkedIn or X. So I used it as a lens, held our current and potential AI practices up against what we'd already committed to, and asked: does this accelerate what we believe in, or does it pull us away from it? In that exercise I didn't find a dramatic gap, but found a lot of clarity.
AI Use Cases Evaluated Through the Charter Lens
AI for Input and Discovery
We collect large amounts of qualitative feedback from students and teachers on our courses that, practically speaking, don't always become actionable due to the sheer volume and the time it would take to meaningfully synthesize. In one recent instance, I pulled that data into a spreadsheet, ran it through AI synthesis, and had a usable slide deck for internal discussion in a matter of minutes.
That is not the same as having a skilled designer or researcher working through the data. But the useful question it surfaced wasn't whether AI synthesis is as good as human synthesis, it was: how do we leverage these tools to scale insight gathering and synthesis without losing the things that make research valuable in the first place? There's a version of principled stewardship that leaves years of user feedback sitting in a spreadsheet forever, and that's not serving our users either. In this case, something usable beats nothing acted upon.
The answer we've landed on is a division of labor that maps well to what AI is actually good at. In many cases, AI can handle (or at least support) the synthesis. We handle the meaning-making: the judgment calls about what the data is really saying, and the advocacy for what the org should do about it. Our charter commits us to understanding and advocating for our users, and this gets us more signal, sooner, with less manual overhead, while scaling human judgment at the point of recommendation across a larger surface area.
AI for Exploration, Clarity, and Alignment
Like a lot of other teams, we've been exploring the use of Figma Make (often combined with planning using other LLMs) to take concepts and make them tangible much faster than before. Increases in speed and volume of output are easy to focus on because they're easy to count, and AI delivers both. But moving faster isn't the same as moving in the right direction. The more important shift isn't speed of output, it's velocity of exploration.
Randy J. Hunt, Head of Design at Notion, offers a useful taxonomy for this (opens in new tab): sometimes you make a design to think, sometimes to validate ideas, sometimes to start a conversation, sometimes to move a conversation forward. Each serves a distinct purpose, and AI makes all of them more accessible, earlier. A faster prototype isn't just a faster deliverable, it's a shared object that lets designers, PMs, engineers, and executives make sure that we're actually discussing the same thing. It's also a learning tool: something you can put in front of real users to test desirability, feasibility, and viability before you've committed to building anything. More options generated faster is not the end goal, but it gets our charter's commitment to reducing risk through exploration and visualization running faster and further than it could before.
AI Maturity is Limited by Design Maturity
Maturity mapping was the most clarifying exercise. When I assessed where my team actually was using Matt Davey's AI Product Design Maturity Model (opens in new tab), our score was low. But what the exercise surfaced wasn't a list of tools we weren't using, it was a set of places where our existing charter commitments had room to run further. Like so many in our industry, my team has held real tension around AI tooling, which I think is healthy and worth preserving. What the charter-as-lens did was reframe that tension: It's not "design vs. AI", it's "are we using every available means to do what we already said we'd do?"
I think this framing matters, because it changes the posture of the conversation inside your team. If you lead with tools, you get anxiety and FOMO and religious debates about which AI tool is worth learning. If you lead with your values and your practice commitments, you get a much more grounded conversation: here's what we believe, here's where these tools can accelerate it, and here's where we need to be careful that speed doesn't cost us something we care about.
Your design charter, or whatever document your team uses to articulate its purpose, is a filter, not just a statement of intent. And if you've done that work already, you have something most teams don't: a definition of success that isn't limited to "we shipped more, faster". Use it.