Empathia AI

Ambient capture, note generation, coding and patient material generation are the whole product, with no services layer or platform business underneath. The company describes a weekly build cycle driven by clinician feature requests, which is a development posture rather than a moat, but it is consistent with a product company rather than a distributor.

Conventional draft and review, with one structural feature that strengthens it: because verbatim dictation is offered alongside AI assisted modes, a clinician can choose a path where the system does not interpret at all for content they want recorded exactly. Notes are finalised by the clinician before use, and live note preview lets them see output forming during the encounter rather than only afterwards. Held at B because no acceptance rate, edit burden figure or confidence threshold is published.

No accuracy figure, model card, named models or evaluation methodology located, and the headline claim of reducing documentation time by over 80 percent carries no denominator or method. One genuine technical disclosure worth crediting against that: dictation is offered in separate VERBATIM and AI ASSISTED modes, which tells a clinician plainly when the system is transcribing what was said and when it is interpreting it. Very few vendors make that distinction explicit, and it is the difference between a record and a summary.

Residency is addressed and the parties are not, which places this mid band. Two deployments exist and they sit in different jurisdictions, with the vendor's own guidance stating that data remains in Canada only if the Canadian version is used while its United States material states storage in the United States.

That is a real disclosure and also a trap, because the protection depends on which instance an account happens to run on, and a buyer who reads one page and signs up through another gets a different answer than they expected. Establish the instance before relying on either statement. On enumeration there is nothing: no model, model family or provider, no hosting arrangement and no sub processor list. The corpus is described only as diverse real world clinical data.

The unresolved question is the boundary around learning. The product adapts to each clinician over time, learning editing patterns, tone and formatting within roughly five encounters, and adaptation of that kind is use of customer derived content by definition.

What is not stated is whether that learning stays inside the individual clinician's own tenant or whether edits and corrections feed anything shared across customers, and those are entirely different propositions for a covered entity. Ask that directly, and ask for a sub processor list per instance.

External validation exists and its nature should be read precisely. Empathia states it is a preferred vendor across multiple Canadian provincial programmes and Canada Health Infoway, which is national body vetting of the kind no United States vendor has an equivalent to, and the same credential held by Tali AI and AutoScribe.

Named site evidence includes an Ontario family health team reporting charting time down by around two hours daily, plus several named physicians in published case studies. The American Academy of Family Physicians offers Empathia as a member benefit, but that is an explicitly commercial discount relationship rather than an evaluation and should not be read as endorsement of performance. No study, controlled comparison or published accuracy benchmark located.

More is established than the earlier pass found, and what it establishes sharpens the question rather than closing it.

Retention is configurable and an instant data deletion option exists. That is a real control and better than a bare process statement. But a health authority's own deployment guidance for this product tells clinicians to toggle instant deletion on and to select the shortest retention option available, which indicates the minimal setting is a choice the customer must make rather than the default they receive. For a product recording encounters of up to two hours, the default matters more than the option, because most buyers never change a default.

The training question is the centre of this axis and it is where the record is least clear. The vendor states its model is trained on diverse real world clinical data, and separately that the product adapts to each clinician over time, learning editing patterns, tone and formatting within roughly five encounters. Adaptation of that kind is use of customer derived data by definition. What is not stated is the boundary: whether that learning is confined to the individual clinician's own tenant, or whether edits and corrections feed anything shared across customers. Those are entirely different propositions for a covered entity, and the marketing language does not distinguish them. Ask directly whether customer content, including clinician corrections, is used to improve models available to anyone other than that customer.

One further scoping point. Two deployments exist, and the vendor's guidance states that data remains in Canada only if the Canadian version is used, while its United States material states data is stored in the United States. Establish which instance a given account runs on before relying on either.

Among the broadest regulatory coverage claimed in this category, spanning HIPAA in the United States and both PHIPA and PIPEDA in Canada, with the vendor describing itself as one of the few scribes fully compliant with both jurisdictions and maintaining separate EMR integrations for each market.

Graded B on the standard applied to multi jurisdiction compliance claims generally: breadth across regimes is real and creditable, but business associate agreement terms are not published for inspection, so a buyer cannot confirm what has actually been agreed in any one of them.

The earlier assessment recorded no attestation located. That is overturned. The vendor states it holds SOC 2 Type II and ISO 27001 certification, and separately published an announcement of achieving the first of those.

Both are genuine independent examinations rather than self descriptions, which is what separates this record from most of the small vendors in this category, and it is a coherent posture for a product marketed as meeting both United States and Canadian requirements.

Also published, and more specific than the tier norm: hosting on the major public clouds within Canada for the Canadian offering, encryption in transit and at rest, multi factor authentication, access controls and audit logs, continuous monitoring and a stated patching practice.

Three things hold this short of the top grade. The certifications are claimed in a frequently asked questions page and a blog post rather than through a trust centre, so there is no dated report, no scope statement and no mechanism for a counterparty to request evidence. Neither claim carries an audit period, and attestations of this kind expire. And the scope question is unusually live here because the vendor operates across three cloud providers and two national deployments, so which systems and which jurisdiction a given report covers is not something a buyer can infer.

Ask for the reports, their periods, and specifically whether the scope covers the Canadian and United States deployments alike or only one of them.

No clearance claimed and none required for ambient documentation. The clinician reviews and finalises the note, and the vendor states that timestamps and audit trails accompany every encounter to support documentation integrity.

What governs varies by market and the vendor operates in two. In the United States no device pathway attaches. In Canada the obligations come from professional regulation rather than a device authority, and the vendor references the national physician mutual defence organisation's guidance and provincial College documentation standards, alongside a recommendation that clinicians inform patients an AI scribe is in use. Referencing the regime that actually applies is better than silence, and it is credited, though the engagement is a line in a frequently asked questions page rather than the published guidance or consent wording that peers in this category supply.

Two boundaries sit around the documentation core. The product suggests diagnostic and procedural codes, described as helping avoid missed billing opportunities, which is the reimbursement boundary and the phrase is doing work: a system oriented toward opportunities missed is pointed in one direction.

The second is patient facing. The product prepares patient education handouts in real time. Material generated by a model and handed to a patient is a different object from a note a clinician signs, because the reasoning that keeps clinical decision support outside device regulation depends on a professional being able to review the basis, and a patient is not a professional. Establish whether a clinician reviews each handout before it is given, and what sources the educational content draws on.

Markets are Canada and the United States. No United Kingdom or European Union presence was located, where ambient documentation is treated as software as a medical device.

Engineers for the language access failure mode rather than disclaiming it, which is the reasoning this index applied to Corti and voize. Multi speaker recording is built to distinguish clinician, patient and caregiver, and the vendor names the specific contexts it is for, including paediatrics, geriatrics and mental health visits where a family member or INTERPRETER is actively participating, so speech is attributed rather than merged.

Transcription runs across more than 30 languages and the product generates bilingual patient handouts, extending language access past the note to what the patient takes home. Held at B because no fairness statement, subgroup analysis or published performance breakdown by language or accent was located, so the design intent is stated rather than demonstrated.

One user facing control here is genuinely uncommon and is the reason for the grade. Dictation is offered in separate verbatim and assisted modes, so the clinician is told plainly when the system is transcribing what was said and when it is interpreting it. That is the difference between a record and a summary, and knowing which one you are reviewing determines how you review it.

Very few vendors make the distinction explicit, and it is a better protection at the point of use than most published accuracy figures, because it tells the person signing where the machine's judgement enters. Configurable retention and an instant deletion option add further customer control. What holds the grade down is that none of it is a commitment about correctness.

The headline claim of reducing documentation time by more than 80 percent carries no denominator or method, no accuracy figure or evaluation was located, and no warranty, indemnity or remediation commitment was found. One deployment detail affects what a buyer is actually agreeing to and should be settled first: a health authority's own guidance for this product instructs clinicians to switch instant deletion on and choose the shortest retention available, which indicates the protective setting is a choice the customer must make rather than the default they receive. For a product recording encounters of up to two hours, the default is what most practices will live with. Establish the defaults and which regional instance the account runs on.

Integration is claimed as tailored per market, with separate EMR connections maintained for United States and Canadian systems, which is more deliberate than a single integration list stretched across both. Held at B because no named electronic record integrations, architecture or certification programme membership was documented or verified in this pass.

Partial, with one genuinely useful property. The product records encounters up to two hours even on low internet connectivity, which matters in rural and remote practice where most cloud dependent scribes simply fail, and is a deployment characteristic rather than a marketing claim.

Against that, no hosting region, residency option or sub processor detail was located, and residency is a live question for a vendor claiming Canadian provincial compliance, where data location expectations are explicit.

Partial. A free trial is offered, plans are described as flexible and affordable with selectable add ons, and a discounted rate is available through American Academy of Family Physicians membership, so a buyer knows the shape of the offer and that a discount route exists. No rate card, tier structure or per clinician figure was located, so none of it can be budgeted from.

Among the broadest enumerated coverage in this category. Specialty pages span psychiatry, paediatrics, oncology with staging and multi modal treatment terminology, ENT including procedure documentation, allergy and immunotherapy, emergency with discharge summaries, and chronic disease with multi problem visits. Clinician coverage extends beyond physicians to nurse practitioners, physician assistants, physiotherapists, occupational therapists, dietitians and social workers.

Encounter coverage is equally wide: in person, a built in telehealth capability handling both phone and video with an automatic note, and recording from third party platforms including Zoom and Doximity. Notes can be organised by problem rather than by visit, and transcription runs across more than 30 languages.