The Borderline co-founder on why she’s arguing with her clients’ Claude documents instead of her clients, and the founder mistakes she sees on repeat.
Julie Mattson’s clients used to send her briefs. Now they send her Claude documents dressed up as briefs. She reads them, starts building the strategy, and finds herself in an argument, except the person she’s arguing with isn’t in the room. “Now I’m arguing with Claude, not them,” she said. That’s the state of marketing in 2026, and it’s exactly why Mattson, co-founder of The Borderline, joined Deconstructing Data to talk about what AI in marketing actually looks like once you get past the demo.
Mattson has been in advertising for 25 years. She started her own ad agency 13 years ago, spent years watching small businesses get priced out of the talent that major brands take for granted, and eventually built The Borderline to close that gap. Along the way she’s watched AI eat parts of her own business, her production and photo shoot work has shrunk considerably, while opening up entirely new ones. Her take isn’t anti-AI. It’s sharper than that: AI is a tool, and tools don’t get to make the decisions.
What does “using AI in marketing” actually mean right now?
Ask ten marketers and you’ll get ten different answers, which is part of the problem Mattson is naming. For her production company, AI has gutted a chunk of the photo shoot business, her creative team can now generate realistic images, including hand motions, that used to require a studio and a full crew. For landing pages and design work, tools like Claude have compressed weeks into hours.
But Mattson draws a hard line between using AI as a collaborator and using it as a decision-maker.
“I’m working with Claude like he’s my designer. I’m not working with Claude like he’s a marketer. And I think if you don’t have that distinction, that mindset going into it, you’re just going to turn things out that look like everything else. I mean, I can spot a Claude deck from a mile away.”
— Julie Mattson, Co-Founder, The Borderline
That’s not a small distinction. It’s the difference between AI speeding up execution and AI quietly setting the strategy, which is where Mattson says most of the real damage happens.
Do consumers actually notice when content is AI-generated?
Yes, according to Mattson, and it costs brands more than most realize. She’s blunt about the psychology at play: people want to believe in what they’re seeing, especially in categories like fashion and beauty, where aspiration is the whole point.
“If I’m wearing, you know, if I shop the runway and I look at all the models, I’m like, I want to look like her. If she was an AI generated model or in a magazine, I’d be less engaged. And I think that’s a big problem with AI generated in marketing.”
— Julie Mattson, Co-Founder, The Borderline
She’s seen it hit closer to home, too. Her mother sent her a product recommendation from what turned out to be a fully AI-generated podcast, three fake hosts having a fake conversation about a real product.
“This whole podcast is AI. Like, it was so good that it hit me. I was like, this is so fake… But my mom doesn’t know the difference. So I have to always reply to her, mom, this is AI. This is AI. It’s not real.”
— Julie Mattson, Co-Founder, The Borderline
For Mattson, that’s not just a UX problem, it’s an ethics problem. Passing off AI-generated content as authentic testimony, editorial or otherwise, crosses a line she won’t let her own agency cross with client work.
Why is letting AI write your business strategy so dangerous for founders?
This is where the conversation gets pointed. Mattson has watched a wave of founders, particularly young ones, hand their entire brand strategy to Claude or ChatGPT and then present it as their own thinking. The problem isn’t that AI is wrong. It’s that AI is agreeable.
“I found out that 61% of all college seniors are starting their own business because they’re scared. And this is very dangerous. Because when you tell Claude, I have this idea, and you go through the idea, it’ll be like, you’re amazing. This is the best idea ever. With no context. And now all of a sudden you have fake validation that your idea is genius.”
— Julie Mattson, Co-Founder, The Borderline
That fake validation loop is exactly why she and her co-founders built The Borderline’s mentorship platform in the first place, giving young entrepreneurs access to humans who will actually tell them when an idea doesn’t work, instead of an AI system optimized to keep them engaged and happy.
Her advice for anyone about to hand a strategy deck to AI: don’t. Use it to organize your thinking, not to generate your thinking. “Don’t let Claude do the strategy work. Give it to me,” she said, only half joking, about clients who show up with an AI-written brand strategy and then get frustrated when a professional pushes back on it.
What are the biggest mistakes founders make when starting a business?
Mattson didn’t hesitate on this one: money.
“Thinking they need money. Trying to raise money. I’ve done this. I keep saying, ‘Oh, got to raise some money.’ No, bootstrap. Bootstrap as long as you possibly can.”
— Julie Mattson, Co-Founder, The Borderline
Her reasoning goes beyond dilution math. Taking outside money early means taking on a stakeholder who wants weekly accountability for a vision they didn’t build and may not even care about. If you’re building something with a social mission, she said, do you really want that mission “ruined by some guy who could care less that you’re helping homeless people and he just wants to see where his money’s going?”
The equity trap compounds it. Mattson is candid that she’s made this mistake herself, giving away equity and advisory shares too fast because someone expressed belief in the idea. Her fix: test-run every potential partner before equity changes hands, and get commitments in writing with real consequences attached.
The bigger picture
What Mattson is really describing isn’t a technology problem, it’s a validation problem. AI is exceptionally good at making people feel confident. It is not good at telling them they’re wrong, and confidence without correction is how bad ideas, bad strategies, and bad business decisions get made faster than ever before.
That’s the through-line of this conversation: AI hasn’t removed the need for human judgment in marketing, it’s raised the stakes on it. The businesses and campaigns that will win aren’t the ones using AI the most. They’re the ones that know exactly where AI’s job ends and a human’s job, with 25 years of pattern recognition behind it, begins.
Want to hear more from Julie Mattson? Check out The Borderline or reach her directly at julie.mattson@theborderline.co.
Ready to bring that same discipline to your data? Visit bdex.com and click “Talk to an Expert” to get started.
Watch the full episode: AI in Marketing and Lessons for Founders
This article was adapted from an episode of Deconstructing Data, BDEX’s weekly podcast on data-driven marketing. Tune in live every Thursday at 4:15 PM Eastern on LinkedIn.
About BDEX: For companies that need clean identity data to power their products, BDEX offers unmatched quality and execution. Visit bdex.com and click “Talk to an Expert” to get started.
