Michael Rowe

Making sense of AI in clinical practice: ATOCP at BOA Congress

talk2 min read

BOA Annual Congress 2026 · ATOCP session · Olympia, London · 24 September 2026

Invited presentation

Overview

This presentation starts by looking at what AI is capable of in clinical and physiotherapy practice, and at four common responses to them, which I call sanctuary strategies: denial, retreat, restriction, and resignation. Each of them leaves the clinician watching something happen to them, so what I’d offer instead is that physiotherapists bring something to the session that AI systems don’t have. The models have no memory of you between sessions, and have no idea who your patients are, so what makes their output useful in a clinic is the context a clinician brings to it: your specialism, your patient population, local services and pathways, how you reason, and what you’ve seen.

The second half turns to patients, who are already using AI to research diagnoses and prepare questions before they arrive at the consultation. And they bring their unique context to the interaction as well. Their own AI agents will carry their history and goals into the consultation, and the talk asks what that does to the therapeutic relationship, and to the MDT, when everyone in the room brings their own contextually-invested AI agents.

Handout

The one-page handout covers the structure of the talk, three concepts it depends on (context sovereignty, distributed cognition, and AI agents), four things to do next, and where to read more. It’s written for any clinician, so you don’t need to have been at the talk to use it.

Download the handout (PDF)

Further reading

The talk grew out of my CSP Founders’ lecture and a webinar for MACP members in May. The idea of context sovereignty is developed at more length in Context sovereignty for AI-supported learning, and Context engineering for educators covers the practical side of giving AI the context it needs. It’s written for educators, but the argument carries over to the clinic.

Slides

Use arrow keys or click to navigate slides. Open slides in full screen

Continue reading

  • Making sense of AI in clinical practice: MACP webinar

    An invited webinar for the Musculoskeletal Association of Chartered Physiotherapists. Explores what AI means for physiotherapy practice across the full information ecosystem of clinical work — from AI-assisted diagnosis and documentation to patient agency and the therapeutic relationship. The central concept is context sovereignty: AI systems work from professional context, and controlling that context is both the distinctive human contribution and the most practical skill for the AI age. The session covers the evidence for AI performance in clinical contexts, strategies for maintaining professional agency, how to actively support patients in using AI well, and how the therapeutic relationship is changing as practitioners and patients develop persistent AI agents.

  • Context sovereignty: building human-AI coalitions in physiotherapy

    The Chartered Society of Physiotherapy's annual Founders' Lecture, arguing that the question facing the profession is not what AI can do that humans cannot, but how practitioners shape the context in which human-AI collaboration happens. The talk opens with the disorienting evidence that AI now matches or exceeds clinicians on diagnosis, empathy, and ethical reasoning, then rejects the defensive 'sanctuary strategies' this provokes. Its central idea is context sovereignty: because language models are stateless, static, and contextless, the professional context a practitioner brings is not a comfort but a structural necessity — and controlling it is the distinctive human contribution. It closes with the coming shift to patient context sovereignty and the work of meaning-making in an ambiguous moment.

  • AI infrastructure for academics: how earlier work makes later requests simple

    I gave an AI agent two YouTube links and one sentence, and it returned two catalogued conference talks with slides, transcripts, and appraisal notes attached in Zotero. The request was short because months of earlier work sat behind it, including a fix from March and a tool from September. This post describes that session and an earlier one, and what they show about the difference between using AI to finish a task and using it in a way that leaves your system more capable than before.