Level 3 taught a team to run AI-assisted contractor work as an operation across many jobs. Level 4 asks the people whose names are on the licence, the owner contract, the pay application and the safety programme to take responsibility for AI across the whole company, and to defend what they decided to an owner's auditor, an insurer, a licensing investigator, a building official, a lawyer in a defect claim and their own superintendents. Every module ends with something one of those readers could pick up and follow.
The first three modules place accountability and record it. Module one explains why office, trailer and field spread responsibility for AI until nobody holds it, traces the chain from the qualifying individual through the design professional and the competent person to the officer who certifies, and names what no tool or vendor can carry. Module two writes the company AI governance programme: scope, a permitted-use inventory, decision rights, stop authority, point people, records and a review cycle, placed inside the quality and safety programmes the company already runs and sized for a two-person remodeller or a four-hundred-person builder. Module three builds the decision record that tells a future owner, an insurer or a defect lawyer why a decision was made, aligns retention with existing record duties and claim periods, and keeps the record alive through staff, project and system turnover.
Module four maps the authorities that bind a contracting business and for which work: six genuinely different state licensing models, adopted codes with their local amendments and the authority having jurisdiction, duties that attach to the work rather than the licence, and two differently built state AI statutes. It says plainly that no contractor licensing board has published AI guidance, and builds a company map that decides what a change means before anyone bids.
Modules five to seven cover what the company buys. Module five assesses a supplier before project data goes near it, on what it does with inputs, who owns project models and outputs, security and change management, tells a real answer from an evasion, treats silence as a finding, and runs a pilot with a written hypothesis, a baseline and stop criteria. Module six reads an AI evidence claim the way an estimator reads a bid, sets out what the published code-checking, BIM, estimating, safety and project-controls studies do and do not support, and designs the company's own acceptance test on its own documents. Module seven writes supplier terms for data use, training, subprocessors, model change, breach notice, deletion, exit and acquisition, places an AI rider beside the owner agreement and the standard documents as drafting practice, and allocates rights in AI-generated renderings where the law is unsettled.
Modules eight and nine handle security and fraud. Module eight puts AI accounts inside a written security programme a contracting business can run, across office, trailer, phone and shared tablet, separates the financing and payment-card duties that are law for some businesses from those that are chosen, and covers monitoring, retiring a trailer computer, supplier oversight and incident response. Module nine assumes a familiar voice or a plausible email proves nothing, follows impersonation to pay applications, lien waivers, bank-detail changes and supplier deposits, and builds a call-back procedure that holds on a Friday afternoon and a plan for the first hour after a diverted payment.
Modules ten to thirteen cover the people outside and inside the company. Module ten reads the signed owner agreement for what it says about tools, the privacy thresholds most contractors fall below, residential contract and cancellation rules in California and New York, and disclosure as a judgement. Module eleven covers AI-written advertising and the licence number, fabricated reviews, renderings and capability claims, the cost of contracting without a licence, and where AI-assisted drafting becomes the practice of architecture or engineering. Module twelve designs verification by someone who did not produce the work and the evidence pack an auditor, insurer or investigator will ask for, drawing on certified payroll and electronic injury reporting. Module thirteen covers hiring tools and the Uniform Guidelines, background reports, biometric time clocks, worker classification and homeowner financing, with notice, correction and human review built in.
Module fourteen closes with leadership: why superintendents resist and when they are right, small reversible trials, sequencing the roadmap from low-risk uses, costs stated with their assumptions, and a report to ownership that says what would stop the programme. The capstone is an AI Governance Programme, Scenario Defence and Ownership Report built on a fictional company. The level ships with a printable workbook and eleven templates, and the examination draws forty scenario questions from a reviewed bank.
Everything here is professional education. It is not engineering, design, legal, insurance or safety advice; it does not replace a licensed design professional's seal, the authority having jurisdiction, a competent person on site, legal counsel, or the executed contract and the law of the state where the work is performed. Completing the level earns an independent educational certificate issued by AI Coalition Network with a public verification page. It is not a contractor licence, a trade or safety credential, a qualifying-individual qualification or a compliance certification, it carries no professional education hours, and it satisfies no licensing, registration or safety-training requirement.