Change management for AI implementation in the NHS: Lessons from CLEARnotes

Over the past year, I’ve had the opportunity to support the implementation of CLEARnotes – an ambient scribing tool (AVT) – within NHS services. While CLEARnotes is a relatively straightforward tool compared to large-scale service transformation programmes, the complexity of introducing AI into clinical settings should not be underestimated.

As those familiar with the National CLEAR Programme know, change management in the NHS is rarely simple. The legacy of fragmented IT systems, layered with both enthusiasm and caution around AI, can make even the most intuitive tools complex to deploy. However, AVTs like CLEARnotes holds tremendous potential – not only for easing administrative burdens, but also for addressing broader strategic priorities.

CLEARnotes directly supports NHS strategic goals, particularly around productivity, workforce wellbeing, and clinical efficiency. It reduces administrative tasks, lowers cognitive load on clinicians and offers more objective productivity gains, such as increased patient throughput. Early pilots have shown up to 14% productivity improvements in outpatient settings — a significant achievement.

Importantly, this level of AI use is low-risk: it doesn’t involve clinical decision-making, and every output is reviewed and verified by a human clinician. That makes it more straightforward to align with emerging NHS and government AI governance frameworks focused on transparency, clinician oversight, and safety. These principles are central to NHS England’s 2025 AI-Enabled Ambient Scribing Guidance, with which CLEARnotes is strongly aligned.

One of the biggest challenges with AI adoption in the NHS is the diversity of systems across and within trusts. Integration is rarely plug-and-play. Our approach has been to start simple, using practical, user-friendly interfaces and then building towards more sustainable, integrated solutions over time.

We’ve found that co-designing with services is essential — adapting the tool to local nuances while maintaining organisational consistency. Strong partnerships help ensure that CLEARnotes remains accurate, reliable, and clinically relevant.

We’ve also learned the importance of clear communication and engagement. Whether it’s reassuring clinicians who are sceptical or setting realistic expectations for early adopters, success depends on stakeholder alignment and clear workflow mapping. The more seamlessly the tool fits into existing processes, the smoother the adoption.

Fortunately, governance guidance is catching up with the pace of innovation. Our long-standing expertise in data security and information governance through the CLEAR programme has made it easier to engage with newly formed, cross-functional AI governance groups in trusts.

Clinical safety has remained our top priority — from iterating outputs to reduce risk to ensuring teams are properly trained in how to use the tool. Operational challenges — particularly those tied to change fatigue and service pressures — are very real, but can be mitigated through thoughtful planning, responsive support, and strong local champions.

Looking ahead, it’s clear the NHS must invest in upskilling the workforce in AI literacy and integration. The pressure to boost productivity must not override the need for ethical, safe, and human-centred implementation. Agility is essential — both in governance and operational delivery — if we want to fully harness the potential of AI while maintaining trust and care standards.

CLEARnotes is not just a documentation tool — it’s a bridge to a more sustainable, efficient NHS. When implemented thoughtfully, it has the potential to support not just productivity gains, but a better experience for both patients and clinicians.

We’re only at the beginning of what AVT can achieve — and I’m excited to see what comes next.