Blog
Thinking at the intersection of AI, behavioural science, and digital health
Writing on digital health, behaviour change, human-facing AI, regulation, and what it takes to build systems that hold up outside the lab.
Right answer, wrong messenger: trust in the source decides whether advice lands
A recommendation can be clinically correct and still fail, because the person receiving it does not find the source credible. Trust in the source, and the design of delivery, sit underneath the five uncertainties rather than inside any one of them.
Read article→The hardest thing to say in a product review is that you do not know yet
Teams give confident answers about AI projects because uncertainty reads as a lack of grip. That is a behavioural mechanism, not a character flaw, and evidence is what fixes it.
Read article→Our best AI evaluator was 92% repeatable, and we rejected it
We built four evaluators to measure one thing and rejected all four. Why the strongest one failed, and why the question is not what score your AI got, but how much confidence the evaluator that produced it has earned.
Read article→Most health AI testing scores single replies, and the systems we ship hold conversations
The AI that holds its boundary for nine turns and concedes on the tenth will pass a test that scores one reply at a time. Risk in conversational health AI accumulates across a dialogue, which is not where most evaluation looks.
Read article→Validated against what? The question to ask before you buy AI that talks to patients
Of the 76 evaluators we use to test health AI, only 16 can be compared against an external expert benchmark. Consistency, detection and discrimination are all different claims, and validated on its own tells you none of them.
Read article→AI associated delusions are the downstream signal, and nobody is measuring upstream
If we only look for the most serious outcomes of unhealthy interaction with AI, we are looking far too late. The harder question is what the early behavioural signals are, and how we would measure them.
Read article→Why GLP-1 providers lose patients early, and where AI can help
As the GLP-1 market gets more competitive, retention will matter more than acquisition. Where patients disengage, why churn is a support problem, and how a behavioural intelligence layer helps providers keep patients engaged and improve lifetime value.
Read article→Where the real value is in AI for GLP-1, beyond the chatbot
When people picture AI in GLP-1 care they picture a chatbot. The chatbot is real and useful, but the value has moved to agents, real-time data, and the gap between engaging with an app and changing behaviour.
Read article→The behavioural layer: why the drug or device is never enough
GLP-1 medicines and devices produce substantial weight loss. Sustained outcomes come from the behavioural layer around them, the support, content, and design choices that help patients change and hold that change.
Read article→Right answer, wrong action: behavioural safety in clinical AI
Patient-facing AI can be clinically correct and still make a patient more likely to do the wrong thing. Behavioural safety is the layer most teams are missing.
Read article→Why the GLP-1 Revolution Has Outpaced Behavioural Science
Why obesity treatment innovation still needs stronger behavioural support and product thinking.
Read article→Why GLP-1 weight loss platforms struggle to scale without behavioural support
What scaling GLP-1 services reveal about behaviour change, retention, and operational pressure.
Read article→When does wellness AI become a medical device?
Reflections on where helpful products edge toward regulated territory.
Read article→Responsible AI Development
Why behavioural science needs to sit inside responsible AI practice, not beside it.
Read article→The real challenge with AI in healthcare is not intelligence. It is safety
A closer look at what makes health AI trustworthy enough for real use.
Read article→CTL Communications has made a strategic investment in Sacher AI
A company update on strategic backing and growth.
Read article→Transforming Behavioural Science with Generative AI: Beyond Content Creation
How generative AI can support more than surface-level output in behavioural work.
Read article→The Transformative Power of AI and Behavioural Science: Meeting Industry Needs with Precision and Privacy
A broader perspective on combining AI with behavioural science in health contexts.
Read article→Behavioural Science and AI: The Emerging Frontier of Digital Health
A view on the overlap between coaching, behaviour change, and AI-enabled health products.
Read article→Top 10 Takeaways from Stanford University's Human Centred AI Report
A response to the themes emerging from the Stanford Human Centred AI report.
Read article→Virtual Health Coaches: An Emerging Role for AI Agents in Patient Care
What conversational AI means for patient support and coaching design.
Read article→The Science of AI in Weight Management: An Evolving Evidence Base
Looking at what evidence needs to support AI in weight management products.
Read article→Incorporating AI and Chatbots: A Must for Digital Health Companies
Why many digital health teams now need a clear AI position and delivery plan.
Read article→Boosting Uptake & Engagement in Digital Health Interventions
Engagement is not just acquisition. It is design, behaviour, and operational follow-through.
Read article→Navigating the Future of Digital Health: Building a Non-Medical Device AI Health Coach Chatbot
What product teams need to think through when building non-device coaching systems.
Read article→The Pros and Cons of Building vs. Buying an AI Health Chatbot
A practical decision lens for teams choosing between internal build and external platforms.
Read article→Advancing Product Development with AI and Behavioural Science
A focused post on using AI and behavioural science together in product development.
Read article→Next step
Facing this in your own team?
Book a discovery call to talk through your patient-facing AI.