The Glucose Never Lies®
Learn more about Grace
The detail behind the main page: how Grace was built, how she is governed, and why she is safe to use. Written for the curious, and the rightly cautious.
For the curious, and the rightly cautious
How Grace was built, and how she is governed
Six questions. The ones a serious reviewer should ask of any AI tool in type 1 diabetes care, answered in plain prose. The small thing Grace is. What she is not. How she was built. How she is governed.
What Grace is
An educational advisor for type 1 diabetes, resting on a locked, version-controlled evidence base curated over more than a decade. She sits alongside your diabetes care team. She is free at the point of use for people with type 1 diabetes, their carers, and the people who support them. She is available outside clinic hours because diabetes does not keep them.
What Grace is not
Not a medical device. Not a clinical decision-support system. Not a replacement for your care team. Not personalised; every figure she gives is a population-average estimate at your total daily dose, and your own correction factor stays with the care team who set it. Not a substitute for the human relationships at the centre of care.
1. How was Grace validated?
Grace runs against a locked evidence base built and curated over more than a decade. Every paper she draws from sits in our Mendeley reference library, exported as a BibTeX file, used as the source of truth. She does not cite from model memory.
Her clinical policies, the cohort framing, the age-band routing, the correction-dose framing, the AID Optimiser positioning, are written down, version-controlled, and published in our open wiki for anyone to read. The retrieval layer pulls from 880 pages of curated content across 7,349 typed chunks, routed by document metadata so the right policy reaches the right question. Phillip Hayes, our Technical Director, built that layer in May 2026.
She is not validated in the regulatory sense of a medical device, because she is not a medical device. She is validated as an educational advisor, against the evidence base she rests on.
2. How were families with type 1 diabetes involved in her development?
I am a paediatric Diabetes Dietitian at Birmingham Women’s and Children’s NHS Foundation Trust, supporting over 300 children and young people with type 1 diabetes and their families. I am also a person living with type 1 diabetes, a husband, and a parent. Grace’s framing, what she covers and how she covers it, is shaped by the questions families ask in clinic, on weekends, and at three in the morning. It is also shaped by the questions I have asked myself.
That is one form of patient and public involvement. We have not yet run a formal service-user advisory panel, and we are open about that. The early review group on the patient and coach side includes Vanessa Haydock and Beth Kelly. On the clinical side, Carmel Smart, Peter Adolfsson, Othmar Moser and others are reviewing against their specialisms.
3. How has Grace been piloted?
Grace has been live to a closed group since late April 2026. As of early June 2026, over 855 people are registered. The Birmingham Children’s Hospital paediatric diabetes operations team have first refusal ahead of public launch, so the team who taught me how to listen to families get to use her before anyone else does.
Named clinicians across the UK, Europe, Australia and the US hold full-access seats and are testing her against their specialisms, from AID and exercise physiology to paediatric dietetics, alcohol and type 1 diabetes, and psychosocial outcomes. The full picture, who is reviewing what and when, is documented openly in our project repository.
4. What is the theoretical approach underpinning her?
Via Negativa. Clarity by subtraction, not coverage by addition. Grace is built by removing what does not serve the central argument, not by adding more reasons until the argument feels safer.
She is an educational advisor. She is not a clinical decision-support system. She gives population-average estimates at a person’s total daily dose. She does not give a personalised correction dose. She refers every clinical decision to the person’s care team. She uses banned-list language to avoid prescriptive framing.
The theoretical frame for what she is for sits in the work of the diabetes psychology and structured-education communities: that the conversations between people with type 1 diabetes and their clinicians become softer ones about them as a person, not just the numbers. Grace is one tool to help make that happen. She is not the whole of the answer.
5. How is the information evidence-based?
Grace only deals with Grades A to D evidence. Grade E content, lived experience, podcast material, FAQ answers, exists in our wiki but is never cited as proof. The wiki backbone is research-grade and feeds her retrieval. The site delivery layer, the voice the reader sees, is evidence-anchored in spirit, not citation-stacked in form. One fully-anchored citation per section is plenty; she carries the depth on demand if you ask for it.
The cohort she rests on is approximately 500,000 patient-days from approximately 1,300 people living with diabetes, run through 77 safety tests across the evidence base. That is the figure used everywhere we describe her, from manuscript to marketing, with no inflation between contexts.
6. What is her mechanism of action?
Grace gets a person 80% of the way there with 20% of the effort. The remaining 20% takes self-discovery, guided by human expertise and trial-and-error learning. That is the strapline, and it is the mechanism we make claims about.
We are not claiming a percentage-point HbA1c change. We are not claiming a percentage-point time-in-range change. The evidence for outcomes like that takes years and dedicated trials, and we are open about not having that yet.
What we are claiming, and where we want the evidence to land over the next year, is that conversations about diabetes become softer ones about the person, not the numbers. That families feel less overwhelmed between appointments. That clinicians have more of their limited time available for the human dimension of care. Those are aspirational claims until the data lands. We say so directly.
Want to hear John walk through the seven-year build in his own voice? Listen to the Tomorrow’s Medicine conversation →
Under the hood
How Grace works, and how the conversation stays private
Question in, graded evidence out, care-team referral always. Every Grace conversation is encrypted in transit, never used to train the underlying model, and linked to a pseudonymous session only. No personal identifier travels with any query.
Built with caution
Is Grace safe to use?
Grace is an educational advisor. She does not give personalised insulin doses. She does not tell you what to do with your glucose. Every numeric output is a population-average estimate at a given total daily dose; the figure that fits your day belongs to the care team who knows you.
Four hard-coded safety rails govern every Grace reply, on every tier, at every time of day.
No prescriptive dose language
Grace never says “take X units”, “your correction dose is”, or “adjust by”. She gives population-average figures at the user’s total daily dose and points back to the diabetes care team who knows the user’s personal correction factor.
Care-team referral on every clinical question
Every answer that touches an insulin decision, a device setting, or a clinical strategy ends with a referral back to the user’s diabetes care team. Without exception.
Age-banded reasoning, not raw year
Clinical decisions route through six bands: infant under 2, preschool 2 to 6, paediatric 7 to 14, adolescent 15 to 17, adult 18 to 64, older adult 65 and above. Grace reasons in the band, not the birthday.
Grade-marked evidence, A to D only
Every clinical claim is graded A (systematic reviews, meta-analyses, large multi-centre RCTs, anchor guidelines) to D (expert opinion, GNL clinical synthesis). Grace signals the grade so the reader can weigh the evidence.
Grace is not a medical device. She is not registered with the MHRA as a clinical decision-support system. She is an educational advisor with a declared evidence base, a declared methodology, and declared limitations. The AID Optimiser sub-tool was reviewed by CamAPS, MiniMed, Tandem and Insulet medical leads; not endorsed by any of them. It is a Grade D synthesis on a Grade A and B evidence base. The IP, compliance documentation, and full safety audit are available to clinical and regulatory reviewers on request: john@theglucoseneverlies.com.
The full framework, in detail
How Learn with Grace maps to the national framework, in full
This is a self-assessment, not a points scheme, not CPD, and not a certificate. There is a UK-wide, four-tier competence framework for diabetes technology (Richardson and colleagues, Diabetic Medicine, 2026; a national consensus statement, endorsed by Breakthrough T1D, the Diabetes Technology Network, Diabetes UK, the DSN Forum, the Royal College of Nursing and Trend Diabetes), operationalised through a self-assessment tool. These bodies endorse the national framework. They do not endorse Learn with Grace, and Learn with Grace is not affiliated with, accredited by, or approved by any of them. That framework is itself a self-assessment, and so is this, so mapping our tiers onto it is a fair, like-for-like comparison.
The framework describes what you understand, not what you are licensed to do. We use the same domains to describe depth of understanding, never to confer clinical competence, scope of practice or permission to act. A strong grasp of the material does not make a non-clinician a clinician, and clinical decisions about real doses always sit with your care team, every time.
Recognise the device, manage the emergency, signpost to the specialist team.
Interpret CGM and pump data, work with the algorithm differences, manage intercurrent illness.
Algorithm internals, advanced interpretation, critical appraisal. We teach the why; hands-on device handling still needs the kit and patient contact.
Service development, publication, mentorship, policy. Shown by doing the work, not by a module set; the continuous, up-to-date surface that opens once all thirty are passed.
One architecture, two cohorts. The clinician framework defines what professionals need to do across four domains: understanding the systems, clinical application, special situations, and signposting. There is no equivalent for the person living with type 1 diabetes. Learn with Grace mirrors those same four domains for the person, not just the professional, so the same map serves everyone.
Each tier is reached the same way: ten modules, each ending in ten rotating questions where more than one answer is plausible, passed at nine of ten. Score low and the path goes deep; score mid and you get a targeted top-up; score high and you go straight to the run.
You leave with a Passport
Every tier you pass is recorded in an exportable Learn with Grace Passport: the modules you completed, the time you put in, your reflections, and where you sit against the four-tier framework. It is yours to download or email; not a certificate, not a licence to act; proof of the work, in your hands.