Aura AI Scribe

Insight Health describes itself as an AI agentic platform and every product in it is model output: the Aura scribe, the Virtual Care Assistant conducting patient interviews, front desk, phone triage and referrals agents. No services layer or platform business underneath.

The scribe itself is conventional draft and review, generating notes and coding suggestions for the clinician to check before use.

Two adjacent agents on the same platform carry materially more autonomy and are not described in oversight terms: a phone triage agent, which triages patients by phone, and the Virtual Care Assistant, which conducts pre visit and post visit patient interviews to capture medical histories unsupervised. No confidence threshold, escalation path or abstention behaviour was located for either.

Graded on the documentation product, which is what this record covers, with the platform autonomy flagged for anyone buying the suite.

Some real mechanical description: automatic speech recognition with speaker diarisation is named rather than implied, note templates carry section level instructions that steer generation, and the white label product documents structured JSON output and event listeners for partners. That is more architectural detail than most. But no accuracy figure, model card, named models or evaluation methodology was located anywhere, so nothing about performance can be assessed.

The boundary is drawn well and no party is named, which is the middle position. On the boundary the published commitments are more complete than most of this tier: a configurable retention period defaulting to twenty one days, destruction performed to a named federal media sanitisation standard rather than described as secure deletion, provider ownership of all patient data, the vendor acting solely as a processor, no sharing with third parties, and no use beyond service delivery.

That last phrase carries the training question by construction, since training on customer content is use beyond service delivery, though the plainer formulation some peers use leaves less room and is worth asking for. On enumeration there is nothing: no model, model family or provider, no hosting arrangement and no sub processor list, despite genuinely useful mechanical description elsewhere including named speech recognition with speaker separation and documented structured output for white label partners.

The white label arrangement itself is a supply chain fact a buyer should hold onto, because it means this engine appears inside other companies' products, so a practice may be using it without knowing the name. One scope question remains open: patient facing agents conduct pre visit history taking, answer inbound calls and triage symptoms, and nothing states whether the retention default and sanitisation standard cover recorded speech from a patient who never attended a visit.

Platform vendor validation rather than clinical study, and the distinction matters. The Epic Toolbox designation in the Ambient Voice Recognition Notes category is a technical verification by Epic that the product follows its recommended connection practices, which is a higher bar than a marketplace listing and is held by very few vendors in this index. The athenahealth Marketplace agreement adds a second platform relationship.

Multiple named clinicians are quoted, including one who reports trialling five platforms before selecting it, and an 11 million dollar Series A closed in April 2026. Held at B because none of that measures documentation quality or clinical outcome: no study, controlled comparison or accuracy benchmark was located, and the vendor claims of two hours saved daily and four to five additional patients weekly carry no denominator.

The disclosures this axis measures are largely present, which is unusual for this tier, and the earlier assessment's concern about breadth remains the right thing to press.

What is published: a configurable retention period defaulting to twenty one days, destruction performed to the United States federal media sanitisation standard, provider ownership of all patient data, the vendor acting solely as a processor, no sharing with third parties, and no use beyond service delivery.

That last phrase carries the training question and largely answers it. Use beyond service delivery is what training on customer content would be, so the commitment reaches it by construction. It is still a use limitation rather than an explicit statement about model training, and the plainer formulation some peers use, saying directly that clinical content is never used to train or fine tune models, leaves less room. Ask for it in that form.

The scope concern from the earlier assessment is what keeps this from the top grade, and the second pass sharpens rather than resolves it. The platform is not a scribe alone. Patient facing agents conduct pre visit history taking by voice or text, answer inbound calls and triage symptoms, and contact patients after procedures to ask about recovery, symptoms and medication adherence. Each of those produces recorded patient speech outside a clinical encounter, and none of the published retention material states whether it is governed by the same twenty one day default and the same sanitisation standard as the scribe.

That is the question to put. A retention policy written for encounter audio may or may not cover a recorded telephone triage call from a patient who never attended a visit at all.

HIPAA compliance stated consistently across product and directory materials. Business associate agreement terms are not published for inspection, and a free tier exists so use can begin before any agreement conversation.

SOC 2 TYPE II is named and typed, which is the top rung of this axis and something only a handful of vendors in this category hold, alongside Tali AI and Scribeberry. The Epic Toolbox designation adds an independent technical review of connection practices, which is a different kind of assurance but a real one. Two additions would make it complete and should be requested at diligence: the audit date and the report scope.

No clearance claimed and none required for the documentation product. No United States device pathway attaches to a note the clinician reviews and signs.

The flag raised in the earlier assessment is confirmed and is larger than a single feature. The platform runs patient facing agents across the whole care episode. One conducts pre visit history taking by voice or text, described by the vendor as capturing critical information just as a physician or physician assistant would. One answers inbound calls, triages symptoms and determines whether the issue is urgent. One contacts patients after procedures at scheduled intervals, asks about symptoms and medication adherence, and flags concerning reports for clinical review.

Determining urgency from reported symptoms is a clinical judgement, not an administrative one, and it is being made in a conversation with a patient where no professional is present. The reasoning that keeps clinical decision support outside device regulation depends on a professional being able to review the basis of an output before relying on it. In a triage call the output is acted on immediately, by the patient, on the basis of what the agent told them.

The vendor's oversight position deserves crediting because it is stated rather than implied. Its leadership describes a safety net for escalation and routing and says a clinician is always in the loop. The questions that follow are the useful ones: what triggers escalation, how a false negative would be detected, and what a patient is told when the agent is uncertain.

The scribe also suggests diagnostic and procedural codes, framed as optimising reimbursement while maintaining compliance, which sits at the milder end of the coding gradient this index tracks.

No fairness statement, no subgroup analysis, no accent or dialect performance disclosure and no language coverage claim were located.

On this record the absence is not principally a documentation quality question, and that is what separates it from every other vendor in this lane. The reasoning needs stating in full because it changes the stakes.

Elsewhere in this category, speech recognition failure produces a worse note. The clinician was present at the encounter, reads the draft, and can correct what the system misheard. The error is contained by the reviewer who witnessed the conversation.

Here the platform's agents speak to patients directly and alone. They take histories before a visit, answer inbound calls and triage symptoms, and follow up after procedures asking about recovery and medication adherence. No clinician is present. The system's understanding of what the patient said is not checked against anyone else's memory of the conversation, and in the triage case it feeds an urgency determination acted on immediately.

So performance across accents, dialects, speech affected by illness, distress, age or hearing difficulty stops being a quality issue and becomes a safety one. The vendor's own material observes that patients calling a practice are often anxious or unwell. That is precisely the speech condition least likely to be well represented in training data, and it is the population the agent is designed to serve.

The disclosure that would matter here is therefore narrower and more urgent than the category norm. Not overall note accuracy, but how reliably the agents understand patients across the range of people who actually call, and what happens when they do not.

Two passes located no accuracy or error figure, no published limitations and no warranty, indemnity or remediation commitment anywhere across the product range. The data side commitments are good and none of them reaches this axis, since a named destruction standard and a twenty one day retention default govern how long content is held rather than whether the output is right. What makes the absence weigh more than for a scribe alone is scope.

The platform runs patient facing agents that take a pre visit history by voice or text, answer inbound calls and triage symptoms, and contact patients after procedures about recovery, symptoms and medication adherence. A triage interaction is a judgement about urgency made without a clinician present, and a post procedure check about symptoms and adherence is a clinical conversation with a patient at home.

Neither has a published accuracy figure, escalation rule, confidence threshold or remediation path. The white label distribution compounds it, because where this engine sits inside another company's product the practice may not know which entity to ask any of this of. Ask for a separate error rate and escalation rule for the triage agent, and establish which company carries the performance representations in your particular purchase.

Among the strongest integration positions in this category and verified by the platform vendors themselves rather than asserted. The Epic Toolbox designation for Ambient Voice Recognition Notes means Epic has confirmed the product follows its recommended connection practices, and the scribe launches directly from Haiku with the encounter captured in context. It is on the athenahealth Marketplace and names eClinicalWorks and AdvancedMD alongside.

The Chrome extension is designed to stay inside the same browser window as a web based record rather than forcing application switching, which is a small but real workflow difference. And Aura Anywhere exposes the whole capability to EHR vendors as a developer product, returning structured JSON that maps to the partner's data model with event hooks and staging infrastructure, which is a deeper integration surface than most competitors offer at any price.

This is among the better answered deployment records in the lane, and the reason is a specificity most of this category does not attempt.

The vendor publishes a default retention period of twenty one days, describes it as configurable, and names the standard used to destroy data at the end of it: the United States federal guideline for media sanitisation. Nearly every peer in this category says data is deleted and stops there. Naming a period a buyer can hold them to, and naming the standard governing how destruction is performed, is materially more useful and far rarer.

Alongside it the vendor states that the healthcare provider retains ownership of all patient data, that it acts solely as a processor, that there is no sharing with third parties, and no use beyond service delivery.

Three gaps hold this short of the top grade. No hosting region is named and no residency option is offered. No subprocessor list was located, and nothing establishes whether a third party model service processes the encounter or what it retains, which matters because a retention commitment binds the vendor and not necessarily its suppliers.

One structural point belongs here. The vendor also sells the scribe as an embeddable component that other software companies install into their own products, delivered through a package manager with the host's branding and single sign on. So a clinician may be using this vendor's scribe inside another company's application without that being apparent, and the retention and processing terms that govern it will sit in an agreement they never saw. Anyone assessing an application that offers ambient documentation should establish who actually built it.

A complete four tier ladder is documented, and it is priced well below the category norm: a free forever plan, Pro at 39.99 US dollars per clinician per month or 480 per year, Team at 79.99 per month or 960 per year, and a custom enterprise tier, with monthly and annual rates both stated. Note that coding suggestions sit in the Pro tier rather than the free one.

Held at B rather than A because the full ladder was located through a third party software directory rather than confirmed on the vendor's own pricing page in this pass; confirming it there would move this to A.

Outpatient focused, consistent with the Epic Ambient Voice Recognition Notes category it is designated in. Specialty fit is achieved through customisable note formats definable per specialty, per practice or per individual provider, with section level instructions steering generation, which is a more granular customisation model than a fixed template library. Named users span physicians and advanced practice providers across several specialties. Held at B because no specialty count, no language coverage and no inpatient or emergency capability was located.