Heidi's 2025 UK Impact Report is not just a product milestone for one AI scribe. It is a signal that UK clinical AI is moving from pilot projects and proof-of-concept evaluations into measurable, operational workflow transformation — at a scale that matters for the health system as a whole.
The numbers are significant: Heidi reports over 15 million UK sessions, an estimated 4 million clinical hours returned, a maximum weekly session count of 414,937, and independent evaluations across NHS trusts, primary care networks, and independent providers. Heidi says it is used by one in two UK GPs and supports 1.8 million appointments per month. These are not pilot-project figures. They describe a technology that has become embedded in the daily workflow of a substantial proportion of UK primary care.
Why This Report Matters
Documentation burden is the most obvious, most universal, and most quantifiable pain point in NHS clinical practice. Two-thirds of clinical staff work additional hours to manage administrative tasks. Twenty-five per cent of NHS medics report burnout, with 20% considering leaving the profession. The Royal College of GPs' survey found that 42% of UK GPs may leave the profession within five years, primarily due to heavy workloads.
Heidi's report demonstrates that ambient voice technology can make a measurable difference to this burden. When a tool reports 4 million clinical hours returned and a 61% decrease in primary care out-of-hours documentation time, the implication is clear: clinicians are getting time back — time that was previously consumed by typing, dictating, and formatting after the patient has left.
What Heidi Appears to Have Demonstrated
Adoption at scale. Used by one in two UK GPs, supporting 1.8 million appointments per month, with 15+ NHS trust partnerships. Modality Partnership — one of the UK's largest GP groups with 53 sites serving nearly half a million people — rolled out Heidi across its entire network. This is not a small pilot. It is large-scale operational deployment.
Capacity release. The report highlights time savings, reduced clinical correspondence backlogs, and increased patient throughput. In the primary care context, a 61% decrease in out-of-hours documentation time means clinicians are finishing administrative work within the working day rather than carrying it home. In independent remote care settings, a 99% reduction in out-of-hours documentation time was reported.
Patient experience. The Modality Partnership deployment reported that over 75% of GPs felt a stronger connection with patients, 100% of patients accepted the technology, and many patients noted better eye contact and more personable consultations. The AI scribe removed the screen as the silent third person in the consultation — allowing clinicians to be present with the patient rather than focused on typing.
Clinician wellbeing. Ninety-five per cent of clinicians in acute settings reported reduced burnout. In Modality's deployment, a similar proportion felt less mentally drained. These are subjective measures, but they are directionally significant — particularly in a profession facing a retention crisis.
Economic value. The report highlights £313,484 in annualised savings in a multidisciplinary community setting, £5.10 return for every £1 spent, and £95,200 annualised value of clinical capacity released in an SDEC pilot. For an NHS under acute financial pressure, the economic case for documentation AI is becoming quantifiable.
Why Ambient Voice Technology Is Now Part of NHS Workflow Thinking
The NHS 10-Year Health Plan, published in July 2025, explicitly identifies AI scribes as a core enabler of digital transformation — calling for the NHS to "scale the use of technology like AI scribes to liberate staff from their current burden of bureaucracy and administration." NHS England's AVT guidance (April 2025, updated April 2026), the AVT Supplier Registry (January 2026), and the AVID community (THIS Institute/Health Foundation) create a national framework for safe adoption. The GOSH-led London-wide evaluation across 7,000+ patients demonstrated increased direct care time, shorter appointments, and enhanced productivity.
This is no longer an experimental technology. It is part of NHS strategy, with national guidance, a supplier registry, and system-level deployment support.
What Comes Next
Heidi's report itself frames ambient voice technology as "the first step in a broader evolution towards clinical decision support and agentic AI workflows." Documentation is the foundation — but it is not the ceiling.
The clinical questions that arise during and after every consultation remain: What does the guideline recommend for this presentation? Which risk score should be applied? Is this medication safe in this patient's clinical context? What red flags should be safety-netted? What should the referral include? What did I learn from this case, and how do I record it for CPD?
These are not documentation questions. They are clinical knowledge questions — and they require a different layer of tooling.
Where iatroX Fits
Where Heidi reduces the friction of producing the clinical note, iatroX focuses on the clinical questions that sit around the note: what the guideline says, what risk score applies, what medicine issue needs checking, and how the clinician can record learning or CPD from the encounter.
Ask iatroX provides cited clinical answers oriented around UK practice. Calculators provide 80+ clinical scoring tools with guideline references. Exam Q-banks support ongoing learning. CPD captures clinical questions as appraisal-ready reflections.
Heidi is helping define the documentation layer of clinical AI. iatroX is focused on the clinical knowledge and learning layer that clinicians need alongside documentation.
Try iatroX for guideline-grounded clinical questions, calculators, exam preparation, and CPD →