Level 2 taught one person to do one dental practice task well with AI, under supervision, with a record. Level 3 is about the operation: the same work run by a team across many patients, providers and sites, where the person accountable for the result cannot personally read every item. It is written for practice managers and operations leaders, group-practice and regional operations staff, revenue-cycle and billing managers, hygiene and clinical operations coordinators, records and privacy officials, and the dentist lead for documentation and imaging.
The first four modules build the operating frame. Module one explains why a task that is safe once becomes risky as a system: volume, handoffs, shared inputs, automation bias on a full schedule and uneven performance across tasks that look alike, and why an operation needs a named owner, a stop authority and a measured baseline before any redesign. Module two maps a front-office, hygiene, documentation or billing workflow as it actually runs, traces patient information against the minimum necessary standard, records who may lawfully do each step under the practice's own state rules and where the dentist must enter, and marks where the claim leaves the practice. Module three tiers AI-assisted tasks by patient consequence, clinical proximity and reversibility, attaches review depth, reviewer, records and escalation to each tier, and explains why an imaging assist sits at the top. Module four writes SOPs a practice manager and dentist lead can enforce, with hard stops written as stops, supervision levels named state by state, radiography permits as a fixed boundary, and a test on a new team member.
Modules five to seven govern the knowledge and the data. Module five builds a retrieval corpus from office procedures, plan documents, fee schedules, forms and dentist-approved patient materials, tags each by payer, state, effective date and code-set year, keeps licensed code content out, and names an owner, a review cycle and a removal process. Module six covers how retrieval fails with confidence, through proposals, vacated or waived text, the wrong year, plan or state, tampered documents and omissions that grow with length, and builds layered defences including a status check made separately from the tool. Module seven limits each role to the information it needs with an access matrix and unique logins, treats radiograph and photograph libraries as their own access problem, explains what each de-identification route requires, and sets authentication and logging that do not create a new store of patient data.
Modules eight to eleven apply the frame to the work itself. Module eight runs an imaging assist across several dentists without letting it become the read: who sees the marks and when, each product's cleared indication as the boundary, acceptance checks on the practice's own images, and monitoring what dentists decided as an operational signal, never a clinical audit. Module nine rolls out AI-assisted documentation as a measured experiment, enforces the draft-until-adopted rule in the system, samples adopted notes against what was done, and states the scope of Texas SB 1188 and California AB 3030 accurately, including that neither names dentists. Module ten audits claims, narratives and procedure mix, treating every flag as a reason to look and never a finding, keeping a small office's audit independent and correcting what an audit confirms. Module eleven runs recall, reactivation and fill-list work at volume, with consent and revocation as an owned operation, the FCC ruling on AI-generated voices, the point where a recall message becomes marketing, and verification sampling.
Modules twelve to fourteen keep the operation honest over time. Module twelve builds a watchlist of the official sources that change dental practice work, from the proposed Security Rule and dated calling-rule waivers to the FDA device list, adverse-event reports, OSHA and the practice's own state board and legislature, and records every check, including those that found nothing. Module thirteen samples AI-assisted work across task types, codes findings with an error taxonomy and severity levels, explains why published dental AI accuracy figures diverge so widely, and turns thresholds and trends into SOP changes. Module fourteen designs escalation routes that hold when every dentist is chairside, stop conditions agreed in advance, measures that would reveal failure, the ways production, acceptance and utilisation measures get gamed, and preparation for the capstone.
Statements of authority are labelled throughout as law or rule, professional standard, professional guidance, best practice, emerging practice, or an AI Coalition Network recommendation, with whom each binds and the date it was checked. What binds the practice is kept apart from what binds manufacturers and payers, and one state's rule is never presented as a national one. The level ships with a printable workbook and thirteen templates, a final examination that draws forty scenario questions at random from a reviewed bank, and a capstone Workflow Redesign of one practice operation built on fictional practices and patients.
Everything here is professional education. It is not clinical training, it does not teach diagnosis, radiographic interpretation, prescribing, treatment planning or how to select a procedure code, and it does not replace the judgment of any licensed dentist or hygienist, legal counsel, privacy or compliance review, payer requirements, or the law and board rules of a practice's own state. Completing the level earns an independent educational certificate issued by AI Coalition Network with a public verification page. It is not a dental or dental hygiene licence, a radiography permit, an expanded-function or coding credential or a compliance certification, it carries no professional education hours, and it satisfies no licensing, permit, registration or payer training requirement.