Question 1

What does the done-for-you claims-allowlist substantiation engagement deliver?

Accepted Answer

Completions builds and operates a 4-skill runtime bundle on the brand-spec agent: Allowlist-Capture, Substantiation-Lookup, Runtime-Validation, Audit-Emission. Skill 1 (allowlist capture) captures pre-approved claims across 10+ regulated verticals: Pharma Rx, Pharma OTC, Supplements (DSHEA), Cannabis (per-state), Food (FDA), Beverage (Alcohol), Cosmetics, Medical Devices, Health & Wellness Services, Financial Services. Skill 2 (substantiation lookup) looks up evidence across 8 source types: clinical study, FDA label, peer-reviewed paper, per-state-AG approved claim, Veeva PromoMats record, MedComms record, RWS record, Aprimo Compliance record. Skill 3 (runtime validation) validates AI-generated output against the allowlist at pre-publish time across 7 check types: exact-match, paraphrase-similarity, entity-extraction, context-coherence, per-state variance, per-vertical-claims-overlay, influencer-claim attribution. Skill 4 (audit emission) emits an audit record with claim text, substantiation ID, validator identity, validation timestamp, decision, rationale, WORM-storage commit, and chain-of-custody. The per-vertical compliance overlay covers FTC substantiation policy + FDA DSHEA policy + state-AG policy + FCC policy + SEC policy + Pharma DTC policy + per-state cannabis-control + per-state alcohol-board + FTC AI-disclosure + FTC Fake Review Rule 2024. Operator owns every artifact: 4 registries in operator data infrastructure; 4-skill model code aligned with operator regulatory-engineering team; per-vendor credentials (Veeva + Aprimo + Workfront + IQVIA + RWS + MedComms + Lionbridge + Bracket) under operator billing; per-vertical compliance overlay rule library in operator repo with attorney-approved updates; per-state-variance register; brand spec; LLM prompts; audit trail. Completions owns the swarm-orchestration knowledge.

Question 2

Why is runtime claims-allowlist gating typically operator-side rather than regulatory-vendor-shipped?

Accepted Answer

The regulated-content management vendors (Veeva PromoMats, Veeva Vault, Aprimo Compliance, Workfront Compliance, IQVIA, RWS, MedComms, Lionbridge, Bracket) ship excellent claim authoring + review workflow + substantiation linking at the content-management layer. The substantiation source vendors (ClinicalTrials.gov, PubMed, FDA Drug Database, DailyMed, NIH NCCIH) ship excellent primary-source data. Neither side ships the runtime 4-skill pipeline that gates AI-generated output at pre-publish time because the claims-allowlist taxonomy reflects operator-specific FDA + DSHEA + state-AG + per-vertical regulatory posture; the per-state variance map reflects operator legal-team approved framework per US state; the per-vertical-claims-overlay integrates with operator legal + medical + finance + creative review workflows; the runtime-validation check semantics (exact-match + paraphrase-similarity + entity-extraction + context-coherence + per-state-variance + per-vertical-claims-overlay + influencer-claim-attribution) require operator-side calibration against the operator-approved validation regime. The work that remains operator-side spans four engineering surfaces: regulatory-engineering for 10+ vertical allowlist capture; regulatory-research for 8-source-type substantiation lookup; NLP-engineering for 7-check-type runtime validation at AI-generation throughput; audit-engineering for WORM-storage + chain-of-custody + 6+ jurisdiction compliance overlay. Completions absorbs all four surfaces under one Tier 3 Fractional CMO with AI Swarm engagement and hands the artifacts back at engagement end.

Question 3

What does the engagement look like across Tier 1 to Tier 2 to Tier 3?

Accepted Answer

Tier 1 AI Readiness Assessment ($10k, 2-3 weeks, diagnostic): audits four axes. Tier 2 AI Swarm Setup Sprint ($25-50k, 4-8 weeks): builds 4-skill bundle on brand-spec agent — completing the 18th 4-skill bundle + Brand-consistency 7-skill cross-swarm mechanic + Pharma/CPG claims-substantiation P19 sub-axis architecture. Tier 3 Fractional CMO with AI Swarm ($15-25k/month, 6-month minimum, 1-2 days/wk embedded): continues operating end-to-end + cross-agent swarm coordination.

Question 4

Who owns the registries?

Accepted Answer

Operator owns 100% of every artifact: 4 registries (in operator data infrastructure), 4-skill bundle model code (operator-owned + operator-regulatory-engineering-team-aligned), per-vendor credentials (Veeva-PromoMats + MedComms + RWS + Aprimo-Compliance under operator billing + operator credentials), per-vertical compliance overlay (rule library in operator repo with attorney-approved updates), brand spec, LLM prompts, audit trail. Completions owns the orchestration knowledge.

Question 5

What measurement and reporting does Completions commit to on Tier 3?

Accepted Answer

Measured against the operator pre-engagement baseline across seven workstreams, reported weekly. (1) Allowlist-capture coverage trajectory — share of operator-approved claims captured into the allowlist registry across the 10+ verticals; per-vertical completeness trace. (2) Substantiation-lookup quality — per-source-type lookup hit-rate against the operator-labeled substantiation set across the 8 source types; per-source freshness trace (substantiation evidence reviewed within operator regulatory team approved refresh cycle). (3) Runtime-validation quality — per-check-type precision + recall + confidence-interval coverage across the 7 check types (exact-match, paraphrase-similarity, entity-extraction, context-coherence, per-state variance, per-vertical-claims-overlay, influencer-claim attribution) against an operator-labeled validation set. Reported against pre-engagement labeling baseline rather than at a 99.9-percent target because validation precision depends on the labeled-validation regime + per-state variance map + per-vertical-claims-overlay rule set operator regulatory team approves. (4) Audit-emission completeness — per-claim WORM-storage commit rate; attestor-identity + chain-of-custody completeness; per-jurisdiction compliance-overlay rule-evaluation completeness. (5) Cross-swarm coordination uptime — per-cycle integration health across the brand-spec swarm. (6) False-positive trace — claims blocked that operator regulatory review subsequently cleared; reported per-check-type against pre-engagement baseline rather than at a sub-1-percent threshold because false-positive rate depends on operator-side calibration of paraphrase-similarity + context-coherence thresholds. (7) False-negative trace — claims published that operator regulatory review subsequently flagged as missed; reported per-check-type with particular attention to Pharma + FDA + state-AG enforcement-priority categories where operator regulatory + legal teams set the acceptable trace ceiling. Process commitments are firm: weekly per-vertical allowlist refresh against operator regulatory team approval cycle; weekly per-source substantiation freshness audit; weekly per-state-variance map sync; weekly per-vertical-claims-overlay rule sync; weekly WORM-storage integrity audit; weekly audit-trail emission. Outcome targets are not promised because validation quality depends on operator-side labeled-validation regime + per-state variance map + per-vertical-claims-overlay rule set + threshold calibration operator regulatory team approves.

Question 6

How does engagement end and what is the operator transition path?

Accepted Answer

Tier 3 engagements are 6-month minimum with 90-day notice. At engagement end, Completions transitions back to operator in-house in 30-60 days: operating-playbook hand-off + in-house staff training + 4 registries hand-off + 4-skill bundle model code hand-off + per-vendor credentials hand-off + per-vertical compliance overlay rule library hand-off + LLM prompts hand-off + audit trail hand-off; Completions credentials revoke immediately.

Frequently asked