Level 3 taught you to run AI-assisted brokerage work as an operation many licensees repeat. Level 4 asks you to govern it across the whole brokerage, and to defend what you decided to a state commission, HUD or a court, an MLS, an errors-and-omissions insurer and your clients. It is written for broker-owners, designated and responsible brokers, managing brokers of multi-office firms, compliance and risk leaders, heads of property management and team owners who own AI policy for a brokerage.
Modules one to four place accountability and map what binds the firm. Module one explains why a brokerage of independent agents, teams and vendors spreads responsibility for AI until nobody holds it, what Texas and California licence law place on the broker, what South Carolina's work-product statute and North Carolina's position place on the licensee, and which decisions no tool, vendor or team leader can hold. Module two writes a governance programme sized to the brokerage: a permitted-use inventory, decision rights and stop authority, supervision, fair housing review, records, internal audit, training and a review cycle, structured with voluntary frameworks that are never presented as law. Module three keeps the decision record: why, who and on what evidence, what a successor broker, an investigator, an MLS, a court or an insurer needs from it, retention aligned with periods rules already set, and a record that survives broker and manager turnover. Module four builds the regulatory map: licence law, fair housing, membership and MLS rules, FTC substantiation, RESPA, antitrust exposure for rental pricing software, the Texas and Colorado AI statutes with their effective dates and enforcers, and documents whose status has changed.
Modules five to seven cover what the brokerage buys. Module five questions an AI supplier on training, retention, provenance and drift, on the listing content the brokerage provides to its MLS, on security and change management and on fair housing testing, tells an answer from an evasion, and decides what to do when a supplier will not answer. Module six reads accuracy and fairness claims by what was measured, by whom, on what data and when, explains why valuation error and steering audits do not generalise freely, substantiates the brokerage's own technology claims and designs an acceptance test. Module seven contracts for the listing-data rights the brokerage actually holds under its MLS warranty, for training and retention, subprocessors, model and terms changes and supplier breach notice, and for exit, deletion and change of control.
Modules eight and nine handle security and fraud from the defender's side. Module eight places AI inside the brokerage security programme using the six CSF 2.0 functions, names AI data and agentic risks, sets authentication, monitoring, supplier oversight and disposal, and builds incident response as SP 800-61 Rev. 3 presents it, with California breach notice as the worked example. Module nine reads what the FBI's IC3 reports about impersonation and closing funds precisely, covers AI voices under the TCPA and fake AI listings, and writes protocols for verifying licensees, people and payment instructions and for the first hour after a misdirected wire.
Modules ten and eleven cover clients and money. Module ten reconciles AI use with representation agreements, gives an honest answer to a client who restricts AI, treats chatbot notice as a brokerage choice while scoping the Colorado and California automated decision duties arriving in 2027, and sets consent controls before an AI voice agent places a call. Module eleven applies RESPA section 8 and Regulation X to partner recommendations inside AI tools, separates marketing services agreements from payment for referrals, covers Regulation B for brokers who refer applicants to lenders and compensation from more than one party, and keeps search from filtering listings by compensation.
Modules twelve to fourteen make the brokerage answerable. Module twelve designs independent verification, a fair housing self-testing programme presented as best practice, and an evidence pack that answers a commission, HUD, a court or an MLS while HUD's proposal to remove its discriminatory effects rule is pending. Module thirteen leads adoption across agents who run their own businesses, reads NAR's 2025 technology survey within its limits, pairs training with emphasised accountability and measures adoption honestly. Module fourteen sequences the programme from low-risk uses outward, meets dated milestones such as Wisconsin's and Colorado's January 1, 2027 requirements, weighs costs and forecasting risk at scale, and prepares the capstone: an AI Governance Programme, Scenario Defence and Ownership Report for a fictional brokerage.
Statements of authority say whom they bind: licensing is by state, the NAR Code of Ethics binds REALTOR® members, MLS rules bind that MLS's participants, and voluntary frameworks bind no one. Where authority is unsettled, the level teaches a method of reasoning rather than a confident answer. It ships with a printable workbook and ten templates, and the examination draws forty scenario questions from a reviewed bank.
Everything here is professional education. It is not legal advice, and it does not replace a real estate licence, brokerage supervision, MLS rules, counsel or the fair housing, licensing, consumer protection, settlement services and privacy laws of any jurisdiction. Completing the level earns an independent educational certificate issued by AI Coalition Network with a public verification page. It is not a real estate licence or a professional designation, carries no jurisdictional education hours, and satisfies no pre-licensing, post-licensing, renewal, fair housing training or association requirement.