Behavioural AI for GLP-1 and obesity care
We help GLP-1 and obesity care teams decide what to build, design how it changes patient behaviour, and test that it works, from first idea to a tested prototype.
GLP-1 and obesity care
In a study of more than 125,000 US adults with overweight or obesity, just over half (54%) stopped within a year. Many of the reasons are behavioural: how people cope with side effects, unrealistic expectations, waning motivation, and the everyday barriers that make lasting change hard.
Source: Rodriguez PJ et al., JAMA Network Open, 2025.
How we work in GLP-1 and obesity care →Who we work with
You run a telehealth service, clinic, pharmacy or weight management programme. You want more patients to stay on treatment, better support between appointments, or a patient-facing AI assistant that is safe to use.
Most start with: Design and deliver or Test and prove
You work in life sciences, medtech or digital health, and you need to decide whether a new product or service is worth the investment before you build it.
Most start with: Decide
01
Find out whether an idea is worth building before you commit the budget. A fixed fee Evidence Phase of 8 to 12 weeks tests the idea using the Before You Build Method, and ends with a clear recommendation to go ahead, rework or stop.
Find out more →02
Turn an idea into a tested prototype or MVP, working alongside your own team. We cover the technical planning, product, design, behavioural logic and conversation design. Your team keeps the production build.
Find out more →03
Show that your AI is safe and that it works. We test it with PromptSafe, evaluate it independently using the FAST framework, and can take the research through to peer-reviewed publication.
Find out more →Our team's experience
22
countries where our team has deployed patient-facing AI
500k+
patients in GLP-1 programmes our team has supported
80
obesity clinics our team's work has reached
27
peer-reviewed publications (ORCID)
Led by Dr Paul Sacher, with 26 years across healthcare, academia and industry, and 49 national and international conference presentations.
Decide
An international GLP-1 care provider had heavy support demand and was losing patients early. We helped it narrow more than twenty AI ideas to the six worth building, with a phased roadmap.
Read the case study →Design and deliver
For Allurion, our team designed the behavioural architecture of personalised AI health coaches for obesity and GLP-1 care, deployed internationally and serving very large patient populations.
Read the case study →Test and prove
A team building a patient-facing AI health assistant asked Sacher AI for an independent view on conversation quality before wider deployment. Manual transcript review had looked reassuring, but simulated testing found weaknesses it had missed.
Read the case study →GLP-1 insight
Dr Paul Sacher joins the GLP-1 Digest podcast with Dr Ashwin Sharma to unpack why patients disengage from GLP-1 programmes, and how behavioural AI keeps them supported and progressing.
Paul played a foundational role in shaping strategy across all three domains and helped us build the teams that now lead our AI, behavioural medicine, and research work. His impact at Numan has been profound and lasting.
Jamie Smith Webb
CTO, Numan
User safety was prioritised and the operational tests to check the interactions and outcomes were really thorough. I recommend them to other health technology innovators.
Grace Gimson
Co-founder and CEO, Holly Health
Their transformative work in creating our behaviour change programme has had astonishing results. Our patients have been able to sustain 95% of their weight loss even 12 months after the programme's end, which is truly impressive considering the chronic nature of obesity.
Benoit Chardon
Chief Commercial Officer, Allurion Technologies
PromptSafe, our AI testing tool
Pre-deployment testing for conversational AI agents, across hundreds of realistic and adversarial simulated conversations. Free trial, no card required.
Latest thinking
A recommendation can be clinically correct and still fail, because the person receiving it does not trust where it came from. How and by whom advice is delivered matters as much as what it says.
Read article →Teams give confident answers about AI projects because admitting uncertainty can look like a lack of control. That is a behavioural mechanism, and evidence is what fixes it.
Read article →We built four evaluators to measure one thing and rejected all four. Why the strongest one failed, and why a score is only as trustworthy as the evaluator that produced it.
Read article →A short call to understand what you are working on, and whether we can help.
