EliseAI improves housing and healthcare efficiency with AI

Minna Song didn't build EliseAI to write poetry or pass bar exams. She built it to answer the phone when a tenant needs a lease renewed or a patient needs an appointment moved, and to do it at three in the morning without losing patience. That's the thread running through her conversation with OpenAI: AI's most useful job right now isn't dazzling anyone, it's absorbing the repetitive, low-drama work that keeps property managers and clinic staff from doing anything else.

Housing and healthcare are both industries drowning in routine contact. Every apartment complex fields the same maintenance requests, the same leasing questions, the same scheduling back-and-forth, over and over, all day. Clinics deal with an equivalent flood of calls to book, reschedule, and confirm visits. Song's pitch is that a language model, properly wired into these workflows, can take that volume off a human's plate and let the human handle the parts that actually need judgment — the tenant dispute, the anxious patient, the edge case nobody scripted for.

What's notable is the restraint in that framing. Song isn't talking about replacing property managers or doctors' front-desk staff wholesale. She's talking about efficiency at the margins, which sounds unglamorous until you remember that margins are where most operating budgets actually live. Housing operators and healthcare administrators don't need a model that reasons like a genius; they need one that doesn't drop calls, doesn't forget context between messages, and doesn't need a coffee break.

It's also a useful data point for OpenAI itself, which increasingly leans on these customer conversations to make the case that its models earn their keep outside chatbots and coding assistants. EliseAI isn't a household name, but it's exactly the kind of quiet, infrastructure-adjacent deployment that determines whether generative AI turns into durable business software or stays a novelty people trot out in earnings calls.