
Artificial intelligence is becoming part of nearly every creative industry, including illustration, graphic design, photography and, yes, puzzles. At Oakland Puzzle Company, our position on AI did not begin as a carefully written policy. It developed through experience, some uncomfortable lessons, and a lot of learning.
At its core, our AI policy is about people: we work with artists whose craft we trust, and we invest in those relationships over time. Artists share their work and their process with us. We pay royalties, put their names and stories on the box, and help their art reach people who might never have found it otherwise. Within that, one line is firm: we do not use generative AI to create the finished artwork for our puzzles.
That does not mean AI-enabled technology never appears anywhere in the process. Many of the digital tools artists use now include some form of AI-assisted functionality. What matters to us is where the creativity comes from, who made the artwork, and whether the finished image represents the work of the artist whose name is on the box.
AI Wasn’t on Our Radar When We Started
We first had the idea for Oakland Puzzle Company in 2018, when generative AI wasn’t really part of the conversation. We don’t come from the art world or the tech world. We were people with an idea, and when we finally got the business off the ground, we were still working from those 2018 assumptions about how art gets made.
Our approach to finding artists was personal. We met people at art shows and vending events, visited galleries, found artists on Instagram and connected through the local creative community. We talked with artists about their work and looked for images that would make an engaging puzzle.
The assumption felt straightforward: if we were talking to an artist about their artwork, the artwork was theirs. We eventually learned we needed to ask more questions.
The Puzzle That Changed How We Think About AI
At a local event, I met an artist whose booth included an image of Lombard Street in San Francisco. It caught my attention right away: great color, interesting lines and the kind of visual detail that makes a puzzle fun to put together.
I asked whether the image was the artist’s work and understood that it was. We licensed it and released it as a puzzle, and customers loved it. It became one of our bestsellers.
Then a customer posted the completed puzzle on Reddit, and several people commented that the image looked AI-generated. Looking into it further, we came to believe the same.
That was the moment everything changed for us. We had inventory featuring artwork that didn’t line up with what we believed we were supporting. It was embarrassing and surprising, and it made clear how much we had to learn.
I’ll be honest: we didn’t pull the puzzle right away, and deciding what to do took longer than I’d like. People sometimes expect a small business to move quickly on something like this. The reality is that we were two partners who didn’t always agree, everyone was wearing several hats, and even important decisions had to compete with everything else on the list. In the end, we labeled the puzzle clearly as AI-generated on our website and discounted it until it sold out, since many customers still loved it. Then we got to work on a policy.
Looking back, there were red flags I would recognize now that I didn’t understand then. I didn’t know how to identify AI-generated artwork, and the puzzle release was rushed. We didn’t have a habit of asking artists the kinds of questions we ask now.
AI has become part of our startup story. It’s not the part I would have chosen, but it pushed us to think much more carefully about artists, creative process and transparency.
Why We Didn’t Adopt a “No AI Ever” Policy
Our first instinct could have been to have a strict no AI policy, banning it from any part of the creative process, but I knew we needed a policy that we wouldn’t have to back down on, and would stand the test of time. I knew we needed something that wrapped around the tools that our trusted artists relied upon. We began to learn more about the role of AI in the everyday lives of artists, and a kneejerk policy didn’t seem useful or realistic.
Many professional creative tools now include AI-assisted features. Most photo editing software on the market relies on algorithms and AI. Our photographers use these tools, and we sometimes use them to create enough bleed for manufacturing.
Artists may also use AI for research and composition planning. To us, that’s very different from using a generative AI platform to create the finished image and presenting it as an artist’s original work.
Our policy isn’t anti-technology. It sets a reasonable boundary around where AI fits in the process and protects the part that matters most: the finished artwork should be the work of a human artist.
Want to know more about the people behind our puzzles?
[Meet the artists who create Oakland Puzzle Company artwork →]
Their backgrounds, styles, experiences and creative processes are a big part of what makes each Oakland Puzzle Company puzzle different.
Digital Art Is Still Human Art
One unexpected challenge of the AI era is that people sometimes assume artwork is AI-generated simply because it was made digitally.
We saw that after releasing Power to the Puzzle. The illustration has a pen-and-ink style: the artist drew the ink work by hand and colored the piece digitally. He spent months developing it, and we still received comments asking why we were releasing AI artwork after establishing our AI policy.
It isn’t AI artwork. It is human-created digital artwork.
That experience showed us that saying a puzzle isn’t AI-generated isn’t always enough. We need to tell more of the story behind the work: sharing sketches when we have them, interviewing artists, and talking openly about training, techniques, inspiration and process. Whenever we can, we want to show how a piece evolves from an initial idea into the image on the box.
Supporting Artists Matters Even More Now

AI has also made us think harder about the role Oakland Puzzle Company can play in supporting working artists.
Many creative jobs that once required an illustrator, designer or photographer can now, in some cases, be approached with generative tools. For artists who have spent years developing their craft, that shift is significant.
Our corner of the art world won’t solve that larger issue, but we can decide what we support. When we license an artist’s work, the artist earns royalties. Their name and story become part of the puzzle, and their work reaches people who might never have seen it in a gallery, at an art show or on social media.
The support goes both ways.
[Larry Hausen] and [Brian Clarke (Les Toil)] have created many of our best-selling puzzles. We just released our second puzzle with Hilary Williams, [Blooming Trails of Año Nuevo], and we’re commissioning new work from [Brenna Daugherty].
Artists like these share their process with us and with our customers, and those ongoing collaborations are what we want Oakland Puzzle Company to be built on.
Trust Comes From Relationships
Identifying AI-generated imagery is getting harder. The odd details that once gave AI images away are becoming less obvious, so visual inspection alone isn’t a reliable long-term answer.
For us, the answer is relationships. Before we license a piece, we make an effort to understand how the artist works, and we are intentionally building deeper relationships with artists whose processes we know. Many of them are wonderfully open about their work and share sketches, works in progress, studio photos and other parts of their creative process.
There will always be some trust involved, especially since some of our most popular puzzles are created digitally. That trust comes from knowing the artist, not from studying a finished image and deciding whether it “looks like AI.”
What This Has Taught Us
I wish we had understood all of this before releasing that first puzzle, but the experience made us much more intentional.
We ask better questions now. We understand our artists’ processes better, we’re building deeper relationships with artists we trust, and we’re thinking seriously about how to share the story of how artwork is made.
We’re still a small company, and we don’t have the resources to document every artist’s process from start to finish. What we can do is talk with artists about how they work before we release their puzzles, and make those conversations a central part of how we choose artwork.
The rule about generative AI matters. But the relationships are the policy: working with artists whose craft we trust and supporting each other over time.
Celebrate the Artists Behind the Puzzle
Every Oakland Puzzle Company puzzle begins with artwork created by an artist whose work we are proud to share. Explore the collection and discover the artists, stories and creative processes behind the puzzles.
Explore Oakland Puzzle Company puzzles: https://oaklandpuzzle.com/collections/all