The Legal AI OS Blueprint closes by naming what an institutional function does next. The Blueprint sets the operating framework, the diagnostic frame, and the progression cadence. What follows is the operating decision the function actually makes — about its own posture, its own progression, and the instruments it chooses to engage.
Three instruments are available to a function ready to act on the Blueprint. They are sequenced rather than alternative: the diagnostic comes first, the engagement frame follows, and the deepest assessment closes the cycle. Functions that skip the diagnostic and engage advisory first typically produce remediation against an incomplete picture; functions that reach the deepest assessment without the engagement frame typically do not yet meet the Band 5 threshold because the operating reality has not yet caught up with the documented intent.
The Free Baseline Diagnostic
The Free Baseline Diagnostic is the entry-point instrument. It is a self-assessment against the eight pillars and the four Maturity Lenses, producing a calibrated composite that places the function on Bands 1 through 4. The Diagnostic is free because the institutional case for moving up the Maturity Stack does not depend on charging at the diagnostic stage. It is calibrated against the Annual Legal AI OS Index so the function knows where it sits against peer functions, not just against an abstract scale.
Outputs from the Diagnostic include a per-pillar maturity score, a per-lens decomposition, a prioritised remediation path, and a referenced peer benchmark. The Diagnostic is the instrument the General Counsel uses to brief the executive committee on where the function actually stands. It is also the basis on which any subsequent engagement decision rests.
Advisory engagement against the 90-Day Roadmap
Functions that the Diagnostic places below Band 4, and that determine moving to Band 4 or 5 is institutionally necessary, engage the 90-Day Roadmap. The Roadmap operationalises the move under disciplined cadence: Diagnose (Days 1-30), Document (Days 31-60), Defend (Days 61-90). Advisory engagement supports the function across the three windows, supplying institutional method and external scrutiny that internal programmes alone struggle to provide. The engagement is sized to the function's scale, sector, and starting position.
The engagement frame is institutional, not technological. It does not replace the function's own operating discipline; it accelerates the function's development of that discipline. Engagements produce the artefacts the function thereafter maintains internally: the Defensibility Posture Statement, the Evidence Register, the AI BoM, the four-quadrant ROAI scoring, the autonomy band determinations.
Executive Diagnostic — independent assessment
The Executive Diagnostic is the deepest assessment and the only tier that determines placement at Band 5 Defensible. Self-assessment cannot reach Band 5. The architecture of separation between self-assessment (Bands 1-4) and the independent assessment (Band 5) is the structural mechanism that makes the Defensibility framework meaningful. A claim of Defensibility that the function makes about itself is not the same institutional posture as a claim a function holds under independent assessment.
The framework defines an independent assessment pathway for Band 5: an independent reviewer (independent of advisory engagement) verifies that the operating reality matches the documented Posture Statement, running the twenty-four-hour Defensibility test against the actual Evidence Register, examining the Five Defensibility Elements as currently operated, and assessing against the four Maturity Lenses, with periodic renewal. These independent certification and attestation capabilities exist within the framework and are not currently offered as part of this service.
For functions not yet ready
Not every function is positioned to engage the Roadmap today. Functions at Band 1 (Foundational) typically require a pre-Roadmap stabilisation phase that establishes the AI mandate, names the executive sponsor, and constitutes the initial governance committee before the Diagnose window can produce a meaningful starting position. Functions in active strategic transition (acquisition, divestiture, regulatory change of significance) typically defer Roadmap engagement until the transition stabilises. The Blueprint does not impose a single cadence on all functions; it names the instruments and the sequence and lets each function calibrate the timing.
What the Blueprint does not do
The Blueprint does not select tools, recommend specific vendors, or substitute for the function's own judgement on operating decisions. It does not pre-determine which use cases the function should pursue; that determination is a Pillar 5 portfolio decision the function makes against its own value thesis. It does not produce a particular ROAI; the framework produces a defensible measurement frame, and the function produces the value within that frame. The Blueprint is Advanta's institutional framework; what the function does with the framework is the institutional posture the function presents to the world.
Where to read further
The full Legal AI OS reference architecture is developed in the anchor essays: Defensibility (the operating framework), ROAI (the four-quadrant return frame), the Risk Taxonomy 2026 (the nine classes of legal AI risk), the Agentic Tier ladder (the autonomy gradient), and the AI Lifecycle (the five-stage operating discipline). The eight pillars are each developed in their own Executive Brief. The Maturity Stack and the 90-Day Roadmap are developed as standalone framework essays. Together they comprise the institutional reference the function operates against.
This assessment determines a function's placement within the Defensibility Framework. Independent certification and attestation capabilities exist within the framework but are not currently offered as part of this service.