Defensible by design
Conceptual, patent-pending claims described in plain language to explain Meridian's defensibility. These are illustrative defensibility concepts, not granted patents.
Defensibility in a regulated domain
Most automation platforms treat the audit trail as an afterthought — a log written when someone remembers to write it. Meridian inverts that. Our approach is audit-by-default: the act of making a decision is the same act that produces its evidence. Because the receipt, the policy version, and the human approval are emitted at decision time and chained together, there is no gap between what happened and what can be proven later.
That matters most in a regulated domain. Fiduciary advice carries books-and-records obligations and supervisory expectations that general-purpose workflow tools were never built to satisfy. A rules-first design — where declarative policy gates every action before it executes — is materially harder to replicate than a black-box model, because the defensibility comes from the structure of the method, not from a particular screen or output that a competitor can simply redraw.
Explainability compounds the effect. When every recommendation ships with the specific rules, thresholds, and client constraints that produced it, the firm can defend the decision, the client can understand it, and a supervisor can replay it. The combination of tamper-evident records, rules-first gating, and generated explainability is the moat: each piece is useful alone, but together they make the end-to-end workflow costly and slow to copy.
What makes Meridian hard to copy
Tamper-evident audit receipts
Every automated action emits a cryptographically chained receipt capturing inputs, the policy version applied, and the resulting decision. Receipts are hashed into an append-only ledger so any later alteration is detectable on replay.
View demo exampleExplainability record generation
Each recommendation produces a human-readable rationale tying the outcome to the specific rules, thresholds, and client constraints that fired. The record is generated at decision time, not reconstructed after the fact.
View demo examplePolicy-gated automated approvals
Workflows advance only when a declarative policy engine confirms every precondition and the required approvals are recorded. Actions that fall outside policy bounds are blocked and routed rather than silently executed.
View demo exampleDrift-to-action workflow with human checkpoint
Portfolio drift is continuously detected and translated into a proposed action set, but execution pauses at a mandatory human checkpoint. The advisor reviews the proposal, evidence, and projected impact before anything trades.
View demo exampleWash-sale-aware tax-loss automation
Tax-loss harvesting candidates are screened against a rolling wash-sale window across linked and related accounts before any order is staged. The constraint check and its inputs are preserved on the decision receipt.
View demo exampleRole-scoped supervision routing
Compliance and risk events are routed to reviewers whose role and jurisdiction match the event, with confidence scoring and full evidence history attached. Escalation paths are deterministic and recorded.
View demo exampleImmutable archival concept
Decisions, receipts, and supporting artifacts are sealed into write-once archival storage on a defined retention schedule. The archive is designed to satisfy books-and-records expectations without manual collation.
View demo exampleNatural-language book queries
Advisors and supervisors can interrogate the book of business in plain language, with every answer grounded in governed data and accompanied by the underlying query and source records. Results are reproducible, not generative guesses.
View demo exampleHow we approach protection
An illustrative view of the principles guiding Meridian's defensibility work. The intent is durable protection of methods, not cosmetic coverage.
Method claims over interface claims
We focus protection on the underlying methods — how decisions are gated, recorded, and made replayable — rather than on screens or visual layouts that are easily designed around.
Regulated-domain specificity
Claims are framed against the constraints of fiduciary advice and books-and-records obligations, where general-purpose automation patterns fall short of what supervisors require.
Evidence as a first-class output
Defensibility centers on producing audit evidence as an inherent byproduct of execution, not a reporting layer bolted on afterward. The evidence path is part of the protected method.
Layered, defense-in-depth filings
Rather than a single broad filing, the strategy contemplates a portfolio of narrower, complementary concepts that together make the end-to-end workflow costly to replicate.
Defensibility that preserves advisor autonomy
For firms
- Lower regulatory exposure: every action carries its own provenance and approval trail.
- Scale service without scaling supervision risk, because oversight is built into execution.
- A defensible moat that protects investment in workflows from fast-following copycats.
For clients
- Transparent decisions: a plain-language rationale accompanies recommendations.
- Trust that automation operates within stated constraints, with humans in the loop.
- Durable records that stand up to scrutiny years after a decision was made.
See the controls behind the claims
Defensible workflow design reduces regulatory risk while preserving advisor autonomy. Meridian's concepts focus on traceability, explainability, and controlled automation — review the security posture, then see it running in the demo.