The AI that reached this yacht brokerage came bundled

The AI product that won this operator's business was never chosen. Jennifer Kerum, co-founder of Anchor Croatia, a luxury yacht charter brokerage on the Adriatic, runs multilingual client conversations through Gemini because it is already inside the email system her team opens every morning. She evaluated nothing. Nine years into the business, AI has reached exactly two of her workflows.

Both of those workflows are horizontal. Every problem specific to yacht charter is still unsolved, and Kerum says plainly that nobody is building for her industry.

Which AI products did the operator name on the record?

Three, and only one of them is a tool she buys.

• Gemini — worked. Translating client email and charter brochures, inside her existing email system. Her reason for adoption was integration, not capability: "Gemini is super easy... It's integrated into our email system."
• ChatGPT — did not work, and she is not the user. It reaches her business through clients quoting it back at her. One asked it whether he could anchor in twenty-five knots of wind, was told he could, and escalated against the captain with a screenshot he had circled in pen.
• Google Translate — named only as the benchmark. Her measure of how far translation has come is what it used to do a decade ago: "You're trying to say, 'Can I please have a coffee?' And it's like, 'I will give you a cat.'"

Naming a tool an operator used is reporting, not endorsement. What matters here is the distribution: the tool that works arrived bundled, and the tool that causes problems arrived through her customers.

Why did the bundled tool win?

Because it was already installed, already trusted and already paid for. The adoption pattern recurs across the operators Artificial Reality, the AI buyer-evidence podcast hosted by Galina Fendikevich, has interviewed: the tool that gets used is the one inside the software the team opens anyway.

That is a structural warning for anyone selling a standalone product into a small operating business. The incumbent suite is not competing on quality. It is competing on already being there, and that competition is not won on benchmarks.

How narrow is the adoption really?

Two functions — translation and marketing content — in a business with roughly a dozen distinct workflows. Kerum's own summary of the scope: "I'd say more on the marketing front and on communication is where we currently use AI in our company."

Neither function was sold to her. Neither required a decision, a trial or a budget line. She adopted them the way you adopt a feature, not the way you adopt a product. And both are capabilities that look identical in yacht charter, dental practices and freight forwarding.

What is left unbuilt?

Everything that is specific to the industry. The problems Kerum describes as unsolved are the vertical ones: matching a client to a boat, remembering what wine a family drank, finding a yacht with light-coloured interiors, and judging which crew makes a charter memorable.

Her account of why:

"You do have some software that is catered to charter companies, but I feel like in general, charter companies and even sales companies and yachting almost get forgotten in the AI world. There isn't a lot of tools that cater to us. And it would just be amazing to have more tools available."

Read that as market structure rather than complaint. The horizontal layer reached her effortlessly because it was bundled. The vertical layer has not reached her because it does not exist.

Is this operator a hard sell?

No, and that is the part worth acting on. Kerum is not a sceptic being talked round.

"That's actually why I'm so excited about AI, because that's kind of our future — that we want to be a yacht company that embraces AI and uses it for our own advantage."

Her read on the brokerages that are sitting it out is unsentimental: "Companies that have been here for a while, I almost feel like they're trying to ignore it, and it's not something you can ignore." She is preparing a separate AI-focused brand, Luxury Yacht Rentals, launching in the coming months. The obstacle in this account is not persuasion. It is that nothing has been offered.

What should a founder take from this?

Two findings, in order of usefulness.

First: in this business, horizontal AI is being adopted effortlessly and vertical AI is not being built. If your product competes for a workflow already touched by a general-purpose assistant inside an office suite, you are not competing against a rival vendor. You are competing against something the buyer never had to decide about.

Second: the industries where nothing is bundled are wide open, and the reason they stay open is not economics. It is that they are physical, seasonal, and a long way from where AI products get built. Fendikevich's version of the builder-side explanation is that Silicon Valley would rather run hackathons and stay indoors than get on a boat.

Kerum's answer to that was an open invitation to come out on one and see the problems in person. So the question for a founder is narrower than a roadmap decision: when an operator with an articulated product need offers you access to her business, what is actually stopping you from taking it?

Where this comes from

S1E2: How a luxury yacht company uses AI without losing the personal touch — the full interview with Jennifer Kerum. Listen or watch: YouTube, Spotify or Apple Podcasts.