Ambient Scribes
H

Hanah

Hanah is an AI medical scribe and patient engagement platform from Melbourne, formerly sold as PhysiPal and cofounded by physiotherapist and clinical lead Aaya and chief executive Ahmed. It drafts notes, letters and referrals from a live consultation or a dictation after the visit, and the clinician checks every draft against its sources before it is saved or sent. Live scribing handles any language, including switches mid sentence, and drafting covers specialist referrals, GP Management Plans, Mental Health Treatment Plans, medical certificates and WorkCover certificates. Visit preparation pulls together prior plans, open problems, recent results and follow up actions, and patient summaries explain terms in plain language.

The scribe runs in a browser or on iPhone, iPad and Android, standalone or inside Cliniko as an official connected app, and in Halaxy, Nookal and Zanda through an extension, with Splose, coreplus, Gentu and PracSuite announced. The same account adds care plans, a patient portal, exercises and outcome measures. Hanah serves allied health, general practice and specialist clinics in Australia, New Zealand, the UK, Europe and the US, with records stored in Australia, the UK or the US. Its October 2026 benchmark of 13 recordings found 93.1% of required details captured and three major unsupported statements.

Practice costs US$80 per practitioner a month, or US$64 billed annually, with a one month trial and no card. Specialist starts at US$200 a month.

AI Health Index verifiedNovember 19, 2026
Compare Hanah with other vendors
Founded
2020
Headquarters
Melbourne, Australia
Website
hanah.health/
Categories
ambient-scribes
Indexed Products
Hanah AI Medical Scribe, Hanah Browser Extension, Visit Preparation, Patient Portal, AI Document Search, Patient Reported Outcome Measures
Buyer Segments
Allied health practices, General practices, Specialist clinics, Solo and group clinicians
Assessment

Capability Axes

An AI Health Index grade measures what a buyer can verify from public sources on the date shown. It is not a rating of how good the product is. A vendor can build an excellent system and grade low on an axis because it publishes nothing an outsider can check. How grades read

AI Capability
AA on AI CentralityThe artificial intelligence is the product. Remove the model and there is nothing left to sell.
Vendor Published

The scribe is the product a practice pays for. It turns encounter audio or a dictation after the visit into a structured note, letter or referral drawn from the conversation and the chosen template, which the clinician then checks. Remove the models and the Practice plan has nothing left to sell. The patient portal, exercises, care plans and outcome measures run on their own, but they make up the free entry tier, which shows that the paid product is the AI. A browser extension and the Cliniko connection put that output into the note workflow of the practice's own system.

BB on Autonomy and Oversight ModelThe oversight structure is described and one part is missing, commonly the threshold at which the system stops or what happens after it is wrong.
Vendor Published

Drafts, letters, referrals and transcripts need clinician approval before clinical use or transfer to a practice system, and the clinician starts ambient capture and confirms patient consent. Inside Cliniko the draft appears in the note form, where it can be edited or declined. The terms warn that generation may produce errors or omissions, and the service logs when a provider safety filter triggers. That is a described review structure. Missing are a threshold that forces escalation when the scribe misses clinical content or writes a plausible wrong detail, and a remediation path once an inaccurate note has been finalized and sent.

CC on Model and Technology TransparencyThe architecture is described in general terms with nothing identified. Proprietary is asserted rather than explained.
Vendor Published

The scribe runs in two stages, speech recognition followed by a separate model that drafts the clinical document, with templates and prompts deciding which document comes back and earlier records feeding document search and summaries. The processing agreement names the supplier roles but not the suppliers, and the speech model, drafting model, versions and training data are all unnamed. Claude grades outputs in Hanah's benchmark, which describes the evaluation rather than the authoring. The benchmark is versioned and dated, so results can be compared over time, but no release log ties a change in note quality to a change in the underlying models.

CC on Model Supply Chain DisclosureThe architecture is described and no model provider is named. Naming a hosting provider alone does not lift a record out of this band. Record the host in the note, because it matters for residency and breach scope, and grade on the model layer, which is the question this axis is named for.
Vendor Published

Hanah's data processing agreement sorts its providers by role, regional hosting, identity, speech recognition and AI drafting, with the processing location for each. Speech recognition is transient, drafting providers may hold text and clinical context for up to 55 days for abuse monitoring, and downstream providers are barred from training on the data. A clinic gets notice before a new processor handles clinical data and can object.

What the agreement leaves out is the legal name of the production transcription and drafting providers. Claude appears only as the grader in Hanah's benchmark, a separate role from writing notes, so the model layer behind the drafts goes unnamed.

BB on Clinical and Operational EvidenceNamed deployments with dated outcome figures and enough method to test them, or published research short of independent validation.
Vendor Published

The October 2026 benchmark ran 13 recorded scenarios against reference transcripts and clinician written checklists. Hanah captured 93.1% of the required facts and drew three major unsupported statement flags across its 13 notes, scored by a blinded Claude grader on a fixed rubric, with significant transcript errors and word error rate measured separately. Scenarios spanned physiotherapy, nursing, dietetics, occupational therapy and a mixed Mandarin and English review.

Hanah ran the test itself, one run per recording, so it shows behavior on those recordings rather than in routine practice. No named clinic reports documentation time, note quality or patient outcomes.

BB on AI Safety and PHI StewardshipCategorical commitments are published, such as no training on customer data, without the retention schedule behind them.
Vendor Published

Clinical data stays out of model training under the agreements between Hanah and its AI processors, and the homepage repeats that patient data is never used to train its models. Speech recognition is transient, drafting providers can keep text and clinical context for up to 55 days for abuse monitoring, the clinic controls record retention, and the terms cover deletion at termination and in backups.

The trust materials add recording consent tracking, safety classifiers and logging of content filter events. Left unstated are whether identifiers are removed before ordinary drafting and how practices are told about a model output safety event, since the breach rules cover personal data incidents only.

Regulatory and Compliance
BB on HIPAA and BAA PostureBusiness associate status is stated and supported by a substantive privacy document, with the agreement or its scope not fully published. For a vendor outside the United States, an equivalent regime documented to this depth grades here. A vendor whose product does not process protected health information grades here too when it says so plainly and explains the scope, with a service privacy document behind it: stating a position a buyer can rely on is the posture this axis grades, and the band above is closed to it because there is no agreement to publish.
Vendor Published

PhysiPal Pty Ltd, the company behind Hanah, acts as the clinic's processor for clinical information under a data processing agreement whose annexes address Australian, New Zealand and UK law, including cross border transfers and the limits placed on model partners. Clinics decide on processing and recording consent, and downstream parties must accept terms at least as protective as the clinic agreement. The trust center maps controls to HIPAA for reference only.

Hanah now offers US data storage and serves US practices, yet no business associate agreement or US tier appears and the US trust link leads to the Australian trust center, so the posture is documented for its home markets rather than the US.

DD on Security Certifications and Trust CenterControls are asserted with nothing independent behind them, or nothing is published. Read the note before concluding anything: this is the grade most often corrected on a second pass, because assurance material frequently sits on a parent domain or inside an old announcement rather than on the product pages.
Vendor Published

Each regional deployment runs on isolated production infrastructure with encryption, role based access and a security risk register, and enterprise accounts can enforce multifactor authentication. The trust center describes logging of sensitive operations and policies for retention and incident response, and maps the controls to the Australian Privacy Principles and the matching ISO 27001 and SOC 2 provisions. That mapping is a self assessment.

No ISO 27001 certificate, SOC 2 report, dated audit or named penetration testing program is offered, and the formal policies themselves are shared only under a confidentiality agreement. The homepage's bank level encryption claim adds no outside check.

BB on FDA and Regulatory StatusThe pathway is stated and in progress, or a clearance is named without the vintage and scope a buyer needs to match it to the product on offer. A product outside device regulation also sits here when the vendor publishes what it is for and in what setting and the product on offer matches, or names another regime it works under, such as a registration outside the United States. A general clause saying the service is not medical advice does not reach this band.
Vendor Published

Hanah sells a documentation assistant that captures the conversation and suggests text for notes, care plans, referrals, certificates and letters, which the clinician reviews. Its terms say the service is assistive and gives no clinical advice, and the homepage states that Hanah does not diagnose or make treatment decisions, leaving diagnosis, treatment and approval of the record with the practitioner.

The company explains this scope for each of its markets and sets out recording consent and privacy duties for Australia, New Zealand and the UK. The product on offer matches that stated purpose and setting, with no device claim and no formal statutory exclusion named.

BB on AI Governance and Bias DisclosureA governance framework with named process behind it, such as certification to an artificial intelligence management standard, or material written for a customer own review committee to evaluate the product with. A framework covering the vendor whole portfolio counts when its scope reaches this product; what it cannot supply is the product specific evaluation the band above asks for. Published results within the one group where the performance of this product is most at risk also sit here: evaluation on the question this axis asks, short of the comparison across groups the band above requires. A medical device clearance or certification, and the quality system behind it, does not reach this band on its own; it is graded under regulatory status. A regulator reviewed plan governing how the model may change does reach it.
Vendor Published

Hanah's benchmark compares each transcript against a reference and checks every generated document for required details and unsupported statements, using checklists written by clinicians and a blinded model grader. Its 13 cases vary noise, specialty and language and report results per recording.

The mixed Mandarin and English consultation produced five clinically significant transcript errors, a published result for the group most at risk for a scribe marketed as working in any language, and a parent with limited English is also in the set. One recording per condition shows where the weak point is without measuring it. No comparison across accents, languages, ages or specialties at scale is offered.

BB on AI Liability and RecourseA published falsifiable commitment, or a real correction route the affected person can exercise against the vendor. A published error rate with its method and denominator grades here. A statutory right that runs to the covered entity rather than to the vendor does not reach this band on its own: every other route here asks something of the vendor, and being located in a particular jurisdiction is not conduct.
Vendor Published

The service agreement warrants reasonable care and skill and compliance with the security measures in the processing agreement, disclaims accuracy warranties for AI output, caps ordinary liability at 12 months of fees and limits the indemnity to qualifying third party intellectual property claims. The clinical terms require review of every draft before use.

Behind that sits a published error measure: the October 2026 benchmark counts unsupported statements and transcript errors with its sample of 13 recordings, checklist method and grading rubric, so a practice can test the scribe's error profile against its own. No obligation to correct a wrong note or remedy a missed detail is attached to the output.

Integration and Deployment
BB on EHR and Interoperability DepthNamed systems with read access or one directional writing, or standards support with named deployments behind it.
Vendor Published

Cliniko lists Hanah as an official connected app, and in Cliniko the clinician records on the patient page while the draft fills sections of the note form. A browser extension brings the scribe into Halaxy, Nookal and Zanda, Splose, coreplus, Gentu and PracSuite are announced as coming, and Hanah describes two way synchronization where a system's API allows it. Practices without a supported system use the standalone app. Those are named integrations with a clear delivery path.

No named practice is shown running the Cliniko read and write connection in production, and for the other three systems the extension writes into forms rather than through a documented record interface.

BB on Deployment Model and Data ResidencyOptions and residency are stated with isolation or the processing path left open.
Vendor Published

Records are stored in Australia, the UK or the US on isolated regional production infrastructure, with New Zealand practices served from Australia. The processing agreement sets out where each step runs for the Australian and UK deployments: clinical storage stays in region, speech recognition processes audio in the United States, and drafting can run in the United States and other countries where the providers operate, with hosting, backups and transfer mechanisms listed for each.

The agreement's annexes cover Australia, New Zealand and the UK, not the US region the homepage now offers. Where drafting runs beyond the named countries and how clinics are isolated within one region are left open.

Commercial
AA on Commercial TransparencyPublished tiers with figures, a stated unit of charge, and a route to start without a sales conversation.
Vendor Published

The scribe is sold per practitioner. Practice costs US$80 a month, or US$64 a month billed annually, and includes AI documentation, document search and summaries, patient retention tools, drafting and staff training, with a one month trial and no card. Specialist starts at US$200 a month for tailored features and dedicated implementation through a quote. The free entry plan covers the patient portal, care planning, exercises and outcome measures without the scribe. Onboarding takes about 10 minutes per clinician, with the scribe usually live in week one, and the terms set renewal, taxes and notice for price increases.

BB on Setting and Specialty CoverageCoverage is named with validation behind part of it.
Vendor Published

Physiotherapy is where the product began, and it now serves general practice, specialist clinics and other allied health teams, solo and group, with document types that follow Australian practice such as GP Management Plans, Mental Health Treatment Plans and WorkCover certificates. The October benchmark covers physiotherapy, dietetics, occupational therapy, myotherapy, community nursing and dictated or noisy recordings, which validates part of that range.

General practice and specialist consultations are thinner in the test set, and a single mixed Mandarin and English review and one parent with limited English stand in for the multilingual use the homepage promises, so validation lags the claimed coverage.

Alternatives

Alternatives to Hanah

Other products that do the same kind of work as Hanah, and what each one does.

Noterro, from Noterro Inc. in Canada, is practice management software for allied health practitioners, built for massage therapy, chiropractic, physiotherapy, osteopathy, mental health and mobile practices.

AI Notes is an Australian clinical documentation app from Black Health Intelligence, an Australian owned company that also trades as TeleScribe.

Lyrebird Health is an Australian ambient scribe founded in 2023, built first for general practice and now extending into specialist, allied health and hospital settings, and selling into the United Kingdom and United States as well.

Sunoh.ai is the ambient AI scribe from healow, an eClinicalWorks company, and the documentation option built into the eClinicalWorks record system.

OneChart is an ambient AI scribe from a small Canadian seed stage team in Hamilton, Ontario. The scribe listens during the encounter, generates structured SOAP, DAP, BIRP and specialty notes, and returns them for clinician review and approval before export.

MedWriter is an AI medical scribe from S7Labs, LLC, doing business as Medwriter, built for outpatient clinicians and strongest in psychiatry, primary care, neurology and pain medicine.

Commercial

Pricing

Vendor-published figures are labeled as such. Figures labeled “Estimated” are derived from third-party sources and have not been confirmed by the vendor.

Entry Price Pricing Basis BAA Tier Implementation Source
$80 per practitioner per month
$80 baseline
Practice, with the AI scribe, costs US$80 per practitioner per month, or US$64 on annual billing. Specialist starts at US$200 a month for tailored features and implementation by quote, and the free entry plan includes the patient portal and outcome measures. The clinical data processing agreement covers clinics in Australia, New Zealand and the United Kingdom. No US business associate agreement is offered. Practice includes staff training, and onboarding takes about 10 minutes per clinician. Specialist adds dedicated implementation support through a quote. Vendor Published

Practice has a one month trial with no card. Plans renew in advance monthly or annually, with taxes applied by region under the service terms.