Lyric

Artificial intelligence is a feature layer on a rules business whose value stands without it. Lyric's security and trust page states that every claim runs on deterministic, rules based logic that is traceable and client controlled, that decision engines run on rules, and that AI surfaces findings and organizes clinical, reimbursement and coding guideline sources so experts can author and apply payment rules faster.

The underlying asset is 35+ years of editing content (the former ClaimsXten) and policy expertise; Lyric42 orchestrates Lyric's own and partner solutions into one recommendation. AI speeds the experts and audit targeting; it does not make the payment decision.

What the system may and may not do are both published, with thresholds and override paths. Lyric's security and trust page says AI does not decide claims or payments: every claim runs on deterministic, rules based logic that the client controls, and AI surfaces findings and organizes source material that a clinical expert reviews every time before a payment rule is written or applied.

The home page says clients configure business rules, review thresholds and decision constraints and can override or tighten any of them at any time, and that recommendations carry their rationale and sources. The automated part is the client's own configured rules. How often edits are overridden on review, and how new AI assisted content is validated before it runs, are not published.

The architecture is described in general terms with nothing identified. Lyric states its decision engines run on deterministic rules, and that AI surfaces findings and organizes guideline sources for experts; Replay applies AI to choosing which claims to audit. No model, technique, accuracy figure or update practice is published for the AI components.

The architecture is described and no model provider is named. Lyric describes deterministic rules engines, an AI assist for experts, and Lyric42's orchestration of partner solutions, and says a vendor and supply chain security program extends its standards to every partner. No partner, model provider or subprocessor is named, and no subprocessor list is published.

Published research short of independent validation. Lyric was named 2025 Best in KLAS for pre payment accuracy and integrity, a buyer survey result rather than a vendor claim, though it measures customer experience rather than the accuracy of edits. The home page reports 35+ years of operation, nine of the top ten US health plans as customers, 10 billion claim lines a year and $20 billion in annual savings realized by clients, which are scale and vendor totals without method. No false positive rate, overturn rate or first pass accuracy figure with a denominator is published.

General assurances that do not answer the questions AI raises. Lyric's security and trust page describes encryption, access controls, zero trust practices and continuous monitoring over the PHI it holds, and HITRUST certifications including an AI security certification. Its home page reports 10 billion claim lines a year and says nine of the top ten US health plans verify claims with it, so the data concentration is very large, and review work reaches medical records.

Nothing published states retention periods, whether one plan's claims experience informs content or models served to another, or what partner solutions reached through Lyric42 retain. Safety engineering for the AI assist is not published.

Compliance is claimed without the document behind it. Lyric's security and trust page describes it as a healthcare technology company entrusted with PHI and reports HITRUST and SOC 2; its privacy notice (effective 13 August 2026) excludes PHI and states no business associate status. The role follows from the work: Lyric processes claims for health plans. Lyric42 routes plan data to partner solutions, and nothing says whether those partners operate as Lyric's subcontractors under a business associate agreement or contract with the plan separately.

Recognized certifications are named without the artifacts. Lyric's security and trust page reports HITRUST risk based r2 certification, the HITRUST AI Security Assessment and Certification, and SOC 2 Type II compliance with independent audits over time, plus penetration testing, zero trust access controls and a vendor security program. No trust portal publishes the certificates or report periods.

No FDA pathway applies and none is expected. Claims editing and payment integrity operate on transaction and billing data with no diagnostic or treatment decision surface.

A real regulatory environment governs this work all the same, and Lyric publishes no position on any of it. Prompt payment laws in most states set deadlines and penalties for claim adjudication, and a pre payment edit that delays a claim sits inside those timelines. Appeal and dispute rights give providers a route to overturn determinations, which makes overturn rates a regulatory as well as a commercial measure. The growing set of state laws conditioning AI involvement in coverage and payment decisions reaches this territory directly.

Unlike utilization management, payment integrity has no accreditation equivalent to URAC or NCQA, so there is no external body examining Lyric's determinations at all. That absence is no technicality. The only checks on a payment integrity engine are the plan's own governance and the provider's willingness to appeal.

Responsible AI is committed to in policy language, with an AI specific security certification and no evaluation of model behavior. Lyric's home page describes 'governed AI' inside a governed architecture, with expert validation, traceable rationale and client control; its security and trust page reports the HITRUST AI Security Assessment and Certification alongside HITRUST r2, and says AI does not make the final call on any claim.

No bias evaluation, false positive rate or appeal overturn rate is published. Payment integrity recommends paying less, and a wrong reduction lands on a provider or member who did not choose the vendor. Where audit targeting uses models, nothing shows whether flag and overturn rates vary by facility type, specialty, practice size or region.

Mechanisms exist and nothing stands behind them. Lyric says each recommendation is verified against policy, codes and documentation and traces to its sources (the policy, the excerpt and the effective date), that clinical experts review what Lyric delivers, and that clients configure and can override every rule, threshold and constraint. That gives a plan, and a provider disputing a reduction, something specific to argue with. Nothing published gives a false positive or overturn rate, an accuracy commitment or a remedy when an edit wrongly reduces a payment; the provider's route is the plan's appeal.

Integration is claimed through an orchestration layer with no system named. Lyric42 is described as reconciling Lyric's edits, reviews and partner solutions into one recommendation through a single integration, which is the right target for a payer, but the company names no claims administration platform, standard or integration method.

A hosted service with the location implied. Lyric's privacy notice says it is based in the United States and stores the information it collects on servers there, but the notice excludes PHI; no hosting provider, region or tenancy model is published for claims data. Lyric has offices in India and the Philippines, so where staff reach claims data from is a residency question too. Lyric42 lets a plan deploy partner solutions through one integration, so claims data can reach companies the plan did not integrate with directly.

No price is published and the posture is discoverable in part. Lyric sells enterprise agreements to health plans across editing, review, enhancement and policy content, orchestrated through Lyric42. Nothing states whether it prices on license or contingency against savings; the distinction changes the vendor's incentive from getting payment right first time toward finding more overpayments. Lyric positions itself on first pass accuracy and less provider friction, yet the pricing model behind each solution goes unstated.

Named and supported: payment accuracy for health plans, before and after payment, across primary and secondary editing, coding review, coordination of benefits, data mining and clinical and reimbursement policy content. Lyric says nine of the top ten US health plans verify claims with it, and it holds 2025 Best in KLAS for pre payment accuracy and integrity. No clinical decisions are made.