AI guides for healthcare and dental practices

These guides are for owners, office managers and practice administrators at dental and medical practices who are deciding whether, and how, to use AI for front desk, billing and insurance work. They are written for the person who has to answer one question before anyone signs up for anything: is this safe to use with patient information?

Who these guides are for

In many practices, AI is already in the building. A front desk coordinator drafts a recall message in a chatbot. A billing lead asks one to explain a denial code. Someone tries an AI phone answering service on a free trial. None of that is unusual, and none of it is a plan. These guides help you turn scattered experiments into decisions you can explain to your team and your patients.

How to think about AI with patient data

Three questions cover most of the ground.

Where does the data go, and who can see it?

Any AI tool that touches protected health information (PHI) becomes part of how your practice handles that information. Before a tool sees a patient name, an appointment or an insurance detail, you should know what it stores, for how long, who at the vendor can access it, and whether you can get the data back or have it deleted. A controlled layer helps: access limited to what each task needs, a log of what was accessed, and a person who approves anything that changes a record.

Will the vendor sign a business associate agreement?

Under HIPAA, a vendor that creates, receives, stores or transmits PHI on your behalf is generally a business associate, and you need a business associate agreement (BAA) with them. A BAA is a contract. It sets out how the vendor protects the information and what happens if something goes wrong. It is not a certification, and no product makes a practice compliant on its own: compliance depends on how the practice uses the tool. Ask each vendor directly whether they sign a BAA, on which plans, and which features it covers. If the answer is no, keep PHI out of that tool.

How will staff actually use it?

Most of the risk comes from ordinary use, not from the software. A clear rule about which tools are approved, what may be pasted into them, and who checks the output before it reaches a patient does more than any feature list. Write the rule down, walk the team through it, and revisit it each time you add a tool. Our AI policy starting points give you a first draft to edit.

What the guides cover

• Dental insurance verification with AI. What can be automated in eligibility and benefits checks, what still needs a person on the phone with the payer, and how to measure the time it takes today.
• Choosing an AI receptionist for a dental practice. Evaluation criteria for phone answering: how calls hand off to staff, what the system can see in your schedule, after-hours rules, and what the vendor keeps from each call.
• Is ChatGPT HIPAA compliant? What the question really means for patient information, what to ask about the plan your staff would use, and how to set a simple rule for the office.
• A HIPAA buyer checklist for AI tools. The questions to put to any vendor about BAAs, data retention, access controls, audit logs and who reviews the output.

Each guide is educational. None of them is legal advice, and none replaces your privacy officer or counsel.

Where to start

Start with the work, not the tool. Count how many hours the front desk spends on insurance calls, confirmations and rescheduling in a normal week. That number tells you which guide to read first and what an improvement is worth. If you would rather have it measured, the Edgerton assessment records the work across participating computers for ten business days, with staff consent, and ends with a report and a proposal. For the practice view, see AI for dental practices and AI for healthcare teams.