where ai does not belong in a small business yet

last week i taped with a property manager out of silicon valley. third generation real estate. she runs other people's buildings and her own portfolio at the same time.

i asked her the question every small business owner is asking right now... where does ai belong, and where does it not belong yet?

her answer stuck with me. not because of where she uses it. because of where she refuses to.

she uses it in two places. maintenance first. a tenant calls in with a problem and the system sorts it before anyone rolls a truck. is it a faulty toaster, or is the whole building losing power? one of those is a shrug. the other is an emergency. you don't want to pay a technician to drive across town and find out which.

then leasing. one listing can pull in a hundred plus inquiries. the machine reads all of them and flags the ones that don't fit... like someone whose move-in date doesn't line up with when the unit is free. nobody on her team has to read a hundred emails to find the few worth a call.

that's the obvious stuff. sorting, screening, triage. work where fast matters more than warm.

then she told me where it stays out. owner onboarding.

the owners she works with are handing her assets that start around two million dollars. she put it simply... "if i'm an owner and i'm going to entrust my property to a property management company... i [have] to trust that that person's going to do a good job."

so she meets them. in person. she walks the property with them. only after that trust exists does the ai come in, and even then it's helping with the reporting, not the relationship.

everyone thinks the question is which tasks ai can do. it can do onboarding. it could write a welcome sequence and answer every question at two in the morning without getting tired.

that's the wrong question.

the right one is which moments are trust moments. the ones where the other person is deciding whether to hand you something they can't afford to lose.

those stay human. for now, and maybe for a long time.

she had another story that shows the same instinct from a different angle. her first short term rental had about four vacant days in its entire first year. four.

not because of some pricing algorithm. because she paid attention to who was booking. they were business travelers. they wanted fast internet, a desk, and a lockbox so they could walk in at midnight without calling anyone. she gave them exactly that.

she knew her customer before she automated anything. that's the order most people get backwards.

i run my own work the same way. the machine does the research, the first drafts, the list building, the reminders. it carries more of the load than i would have believed a couple of years ago. but the first real conversation with a founder... the moment they decide whether i'm someone they want in their corner... that's me on a call. no agent, no sequence.

because "there's a trust that has to be built there," as she said. and trust built by a machine reads like a machine built it.

long term, i think this is where the split happens. businesses that automate everything will be cheap and forgettable. businesses that automate everything except the handshake will be cheap to run and hard to leave.

so if you're trying to figure out where ai doesn't belong in your business yet, try this. write down every moment where a customer decides to trust you with something big. their money, their property, their reputation. keep those human. hand the rest to the machine and don't feel guilty about it.

i wrote up how i split that work, task by task, in the ai method.

so which moment in your business is a trust moment... and have you already handed it to a bot?