Clinician in the Loop AI and Clinical Uncertainty Expert Panel

We certainly need human-centred dialogue to enable sustainable applications of technologies in healthcare. Several technologies always present with a lot of promise but stall in integration and scale in health systems. Currently, #AI is the technology in the news. How do we ensure that #AI & #ML tools can stay integrated in healthcare and augment the intelligence of a system fundamental to our continuous existence on Earth? Come May 8, 2026, we will be addressing this with an attempt to identify common decision points that move beyond the technology itself and focus more on the users of the technology. We look forward to having you there! Thank you UZH.ai, UZH Digital Society Initiative, and UZH Population Research Center for supporting this timely Expert panel. Together let’s build AI Technology rooted in the context it is to be applied in. #clinician #in #the #loop #dsi #prc #uzh

An independent evaluation of the NHS AI Lab, led by The University of Edinburgh and supported by Arden & GEM, explored the impact of one of the world’s first programmes dedicated to accelerating safe and effective AI adoption across health and care. Launched in 2019 by NHS England and Department of Health and Social Care, the NHS AI Lab brought together infrastructure development, research, regulation, ethics and real-world deployment, backed by £143.5 million in funding. The findings show the Lab played a pivotal role in coordinating AI development, testing, adoption and regulation—helping position the NHS as a global leader in evidence-based, responsible AI innovation. Read the full case study here: https://lnkd.in/eQhYusK2 #NHS #HealthcareInnovation #AIinHealthcare #DigitalTransformation #NHSLeadership #HealthTech Sheena Dungey Ray Foley

Over the past year, we have carefully considered whether — and how — life sciences companies can use AI tools in their clinical development for medical products and regulatory workflows. From analyzing clinical data to preparing INDs, NDAs, and BLAs, FDA’s January 2025 draft guidance, "Considerations for the Use of Artificial Intelligence to Support Regulatory Decision-Making for Drug and Biological Products," offers an important framework and raises practical questions about risk and regulatory scrutiny. Read the full article below for our key takeaways on real‑world AI use cases and how to align AI adoption with FDA expectations: https://lnkd.in/eab8Jp_P To learn even more, join us for our fifth annual AI Health Law & Policy Summit in Washington, D.C., on May 13-14, where panelists will explore these and other rapidly evolving health care AI regulatory concerns: https://lnkd.in/eTqSJAtm Authors: Robert F. Church, Elizabeth Jungman, Blake Wilson, J.D., M.S., Heidi Gertner, Kelliann Payne, Melissa Levine, Will Tenbarge, Ashley Grey, JD, MSc #FDA #AI

Responsible AI adoption in healthcare requires strong foundations in safety, trust, regulation, and governance, ensuring that AI solutions are integrated into clinical workflows in ways that create real and sustainable value. This was one of the core themes running through the AI in Healthcare: Leading Responsible Adoption at Scale programme at Imperial College London, Imperial Executive & Professional Education, which I recently completed. As part of the practical component, I explored a proposal for a Human-Agent Team (HAT) Triage System for Emergency Departments, drawing on studies from Harvard University (2026), Sentinel (Kim et al., 2025), and the Nadaf RCT (2025). The proposal examined how AI agents could become part of the clinical triage team by surfacing relevant patient history, flagging risk indicators, and structuring clinical handovers, while clinicians remain fully in control of final clinical decisions. The work touched on regulatory considerations under EU MDR and the EU AI Act, risk and ethics frameworks, AI governance, and a phased implementation approach for safe and controlled adoption. The programme also provided an opportunity to explore safety and regulatory frameworks for AI adoption within the National Health Service (NHS UK), covering governance, clinical product safety, and responsible deployment considerations. Thank you to Brendan Delaney and Luca Parisi for such a well structured and insightful learning experience. #AIHealthcare #ResponsibleAI #EUAIAct #EUMDR #AIGovernance #SaMD #ImperialCollegeLondon https://lnkd.in/d6N6vEZ6

Artificial intelligence is rapidly reshaping clinical practice, bringing both opportunity and risk. At the NPAA & AUNP Conference 2026, plenary speaker Anastasia Semenova will examine how AI is being applied in NP practice and what it means from a legal and risk management perspective. 🧠 Artificial Intelligence (AI) in NP Practice: Legal and Risk Management Considerations 📅 May 29 | 1:45–2:45 PM This session will explore: • Privacy, accuracy, and informed consent in AI use • Professional liability considerations for NPs • Practical strategies for integrating AI responsibly into clinical workflows Grounded in real-world case studies, this session is essential for practitioners navigating emerging technologies in healthcare. NPAA AUNP 2026 | NPAA | AUNP | NP | Alberta Nurse Practitioners | Health Care AI

We often hear that having a “human in the loop” makes AI in healthcare safe. I’m not so sure it’s that simple. In our recent paper in lancet publishing group, we argue that this idea can end up being more reassuring than actually protective. There are a few reasons why. AI systems don’t just make isolated decisions — they can scale existing inequalities in ways that are hard to see and even harder to stop. The kinds of harms that emerge, especially at the intersections of race, gender, class, and more, don’t fit neatly into oversight models that assume a single, neutral reviewer. And in reality, clinicians are working under time pressure and constraints that make it difficult to really question or unpack what an algorithm is telling them. Using perspectives from actor–network theory, feminist epistemology, and Iris Marion Young’s work on responsibility, we try to show how current approaches place too much responsibility on individuals, while letting institutions off the hook. So what could we do differently? We suggest thinking about AI as one input among many in clinical decision-making — not something that quietly becomes the authority. I also think communities affected by these systems should have a real say in governing them, including the power to push back or shut them down when needed. And importantly, responsibility should sit more clearly with the organisations building and deploying these tools, not just the clinicians using them. If AI is going to be part of healthcare, we need to take accountability seriously — not just in theory, but in how these systems are actually designed and used. Great thank to our mentor in this project Leo Anthony Celi and all our colleagues https://lnkd.in/dzZCv4zy

OpenEvidence, the AI-powered medical information service, has recently made the decision to block access to their platform in the UK. This move comes as a direct response to concerns surrounding the UK's evolving AI legislation and regulatory landscape. On one hand, it’s encouraging to see the UK government taking proactive steps to establish frameworks that aim to protect society and ensure AI is developed and deployed responsibly. Safety and ethical considerations must be paramount in this rapidly advancing field. However, OpenEvidence's withdrawal highlights a critical tension. While robust legislation is essential, it shouldn’t inadvertently stifle innovation or create an environment so complex or risky that valuable tools are withdrawn from the market. Medical professionals in the UK are now losing access to a resource that could potentially aid in clinical decision-making. We need to strike a delicate balance. Regulation must be clear, proportionate, and supportive of technological advancement, ensuring that both developers and users can benefit from AI without compromising safety. We must build a regulatory environment that attracts, rather than deters, providers from making their products available in the UK. #AI #HealthcareAI #OpenEvidence #UKLegislation #TechRegulation #Innovation #DigitalHealth #AIethics #NHS #PrimaryCare

💡AI Fairness in Health: Insights from the Better4U Consortium Workshop 🩺 Nearly half of people in Europe live with overweight or obesity. When AI is used to support their health journey, safeguards must be put in place to ensure its fairness. As part of the EU-funded BETTER4U project, our team at the University of Vienna leads Ethics, Legal and Regulatory Compliance. We recently facilitated an internal AI Fairness Workshop with the full consortium, bringing together researchers, clinicians, and data scientists to examine BETTER4U's AI model from a fairness perspective at every stage of its development. 🧑⚕️ What does AI fairness mean in practice? Who could be harmed if we get it wrong? And how do we measure it? Read the full interview to find out. 🔗 More info: https://lnkd.in/dDQJTapk #BETTER4U #EUResearch #HealthTech #DigitalHealth #Research #Innovation #AIFairness #TrustworthyAI #AIEthics

I very much look forward to this discussion on the future of AI, healthcare, liability and regulation with Gordon Wishart, Dr. Gunnar Sachs and others. Thank you for hosting Jonathan Kewley and Clifford Chance

AI has the promise to revolutinise healthcare - through diagnostics, drug discovery, and turbo-charging admin and operational tasks that ensure it is delivered. It is also a democratising force, as anyone can use it learn about conditions, treatments, and how to navigate complex healthcare systems. Join us to discuss how we realise the promise, and manage the risks, of AI in healthcare 👇