AI in general practice: What GPNs need to know

Artificial intelligence offers a range of potential benefits in healthcare, including more accurate prediction of patient risk, improved disease prevention and control, and earlier diagnosis of conditions, but its use should be tempered with caution

Artificial intelligence (AI) tools are increasingly shaping the way we live, work and learn. In healthcare, this shift cannot be ignored. AI is being used in primary care in a variety of ways, and usage is likely to increase as tools become more accessible.

Most importantly, patients are increasingly using AI tools to self-diagnose and guide their care. A recent poll found that around a third of adults in the US turn to AI for health information and advice, especially for mental health concerns.¹ Worldwide, 75% of clinicians report that patients arrive at appointments ‘AI-informed’.² Of concern, 35% of adults who regularly use AI chatbots for health advice believe common vaccine myths, compared with 20% of those who do not, highlighting why general practice nurses (GPNs) must remain a trusted, specialist source of accurate health information.³ If patients are already using large language models (LLMs) such as ChatGPT, Claude, Gemini, and Copilot, healthcare staff must be able to both understand and respond to their impact.

AI offers a range of potential benefits both generally and in healthcare, including more accurate prediction of patient risk, improved disease prevention and control, earlier diagnosis of conditions, enhanced patient empowerment, and reduced workload for healthcare professionals.⁴ At the same time, its introduction brings significant ethical and regulatory considerations, and healthcare professionals must reflect on how these technologies should be used responsibly by clinicians and patients alike. This article focuses on AI in general practice and how GPNs can prepare for its implementation in our day-to-day practice, while maintaining high standards of patient care and adhering to the NMC Code5 and MHRA guidance.

The current picture

AI is generally defined as technology that emulates human intelligence, using a range of algorithm-based technologies to solve complex tasks.⁶ Digital tools, including AI, are discussed in the NHS ten-year plan, which states they will empower patients and 'liberate staff from bureaucracy and administration'.⁷ The NHS AI Lab roadmap predicts that general practice will be one of the groups most affected by the introduction of AI,⁸ because general practices have historically led the way in patient-facing digital services, and their records serve as patients' primary records of care.⁹ Nevertheless, for those of us working in general practice, it is clear that significant improvements are still required to ensure these technologies are fully effective and fit for purpose. Many of us will understand the frustration when the system 'goes down' and alongside conversations about AI implementation, important discussions need to take place to ensure that basic infrastructures are reliable.

A policy paper produced by the last government set out the scale of change needed for healthcare to match the digital maturity seen in other sectors.¹⁰ The report highlights a common frustration in general practice: key documents, reports and test results are not always shared across patient records, making it difficult to access information from other providers that may affect patient safety. Some would argue that the NHS must fix these basic systems before introducing more advanced technologies.

There is considerable variation in clinicians' readiness to use digital tools in general practice, with implementation shaped by complex and multifactorial influences.¹¹ Nevertheless, the integration of AI into general practice is happening right now and GPNs must be prepared to engage with it safely and effectively.

In 2025, the RCGP reported that more than 1 in 4 GPs reported using AI tools in their work, a figure that has almost certainly increased since then.¹² GPs in this study wanted AI to be used to handle 'routine, time-consuming tasks reliably'. There is currently no similar research on whether GPNs are using AI in practice, but given the pressures in general practice nurses, it is reasonable to assume a strong appetite for workload-reducing tools.

The Care Quality Commission (CQC) has a GP mythbuster on the use of AI in general practice, which sets out the benefits, risks and ground rules of using AI in clinical settings. Benefits include enhanced clinical efficiency, reduced administrative burden allowing clinicians to have more meaningful contact time with patients, improvements in patient access to services, cost savings and better data management.¹³

AI is also being applied to some of the administrative problems that quietly drain primary care. Did-not-attends (DNAs) are a persistent and costly issue, with patients missing 16 million GP appointments in 2025 alone.¹⁴ AI tools are now being used to predict which patients are most likely to miss appointments and to tailor reminders or offer more convenient slots accordingly. In a pilot scheme, DNAs were by almost a third in 6 months.¹⁵ Many GPNs will be acutely aware of how frustrating and wasteful repeated DNAs can be, and this represents a promising use of AI to target a real, everyday problem.

What could AI mean for GPNs?

Ambient Scribing and Face-to-Face Care

An important question to ask is how AI could benefit both our practice as GPNs and improve patient experience. Using tools such as ambient scribing could arguably have the biggest impact. Ambient voice technologies (AVTs) such as Heidi and Surgery Intellect are already in use across a number of practices, automatically transcribing consultations directly into patient notes. For many clinicians, this kind of tool represents exactly the sort of timesaving, low-risk application of AI that could make a tangible difference to daily workload.

Research shows that ambient scribing tools can reduce appointment length by 8.2% and increase patient-clinician interaction by 23.5%.¹⁶ Patients also responded positively to the use of ambient voice technology, with many noting improved engagement during consultations. Clinicians using these tools described AI-scribing as 'transformative'.¹⁷

However, this efficiency could easily be absorbed into increased capacity in our clinics: the same research shows 13.4% more patients can be seen per session.¹⁸Time saved by AI should arguably support clinicians rather than increase patient load. Protecting this time is vital for staff wellbeing, GPN recruitment and retention, and the face-to-face care that drew many into nursing. With almost half of clinicians using AI reporting less work-related stress,³ these tools could help reduce burnout if the time they release is protected for those delivering care.

Supporting learning and everyday admin

In general practice we're expected to be specialists in everything, from immunisation schedules to the latest NICE guidance, often across multiple clinical areas. Policies, protocols and training requirements shift constantly, yet GPNs rarely have protected time to keep pace with them. This is where AI tools could offer real value, as an accessible, self-directed way to support our ongoing learning and understanding, fitting around clinics rather than competing with them.

Used correctly, LLMs such as Claude and ChatGPT can summarise guidance and critique research. You can request quizzes on specific clinical topics to test your own knowledge or instruct the tools to create spreadsheet trackers to support evidence gathering for NMC revalidation. You can also use trackers to ensure everyone's training is in one place when required, e.g., in the case of a CQC inspection. However, for these tools to be effective, GPNs must have proper guidance and training on how to use them appropriately and safely.

In my own practice, I have used AI to create a spreadsheet for tracking vulnerable elderly patients, including vaccine eligibility, housebound status by postcode, and QOF or annual review monitoring. However, once identifiable patient data is added, AI must not be used, as this would breach confidentiality and data protection requirements. This highlights the need for secure, closed-system NHS AI models, where data stays within the organisation and is not used to train the model. NHS England’s 2026 rollout of Microsoft 365 Copilot to more than 500,000 staff is encouraging,19 but it is focused on trusts, leaving unclear when general practice and GPNs will gain access to the secure tools they need.

Regulation of AI: what GPNs need to know

Regulation is working hard to keep pace with the technology, and both clinicians and the public have noticed this shortfall. More than three-quarters of healthcare professionals are unclear about who would be liable in the case of AI-driven errors,²⁰ while 82% of patients and the public believe the current regulatory framework is insufficient for data governance and privacy.²¹ Currently, most AI tools used in general practice, including ambient scribing products that generate summaries, are categorised as Class I medical devices, which require only self-certification by the manufacturer and registration with the MHRA.²² Importantly, MHRA registration is a declaration by the company, not an independent assessment of the product.

Registration alone does not make a tool ready for clinical use, however. Before deployment, the practice itself must complete a clinical risk assessment under the DCB0160 safety standard, overseen by a Clinical Safety Officer (CSO), a registered clinician trained in clinical risk management. CSOs are commonly employed within large hospital trusts but rarely within general practice, where this support is usually accessed at integrated care board (ICB) level and is currently in high demand. The CQC has made clear that inspectors may ask practices to evidence this governance during inspection,¹³ making AI procurement a clinical governance responsibility, not simply an IT decision.

In June 2026, the ambient voice technology tool TORTUS became the first of its kind to achieve Class IIa certification, meaning it has been independently assessed by a UK Approved Body rather than self-certified.²³ The MHRA's National Commission into the regulation of AI in healthcare is also due to report in September 2026, and is expected to reshape how these tools are regulated.²⁴

Global examples and looking ahead

There are some genuinely exciting examples of AI being used clinically across the world. These range from robotic-assisted surgeries significantly reducing recovery times,²⁵ to AI-assisted diagnostic tools such as Cerviray AI supporting gynaecological care, where Kim et al found that combining the impressions of the clinician and the AI improved diagnostic accuracy in colposcopy.²⁶ AI is also expanding access to mental health support through platforms such as Wysa and Talkspace, which offer patients affordable and accessible care while easing pressure on over-stretched clinicians.²⁷ Closer to home, researchers are training an AI model called Foresight on de-identified NHS data from 57 million people, with the aim of predicting ill health early enough to intervene, including recognising those with rare conditions and addressing health inequalities.²⁸

Internationally, countries are looking to AI to relieve pressure on primary care infrastructure. Portugal's SNS 24 helpline (the equivalent of NHS 111) became mandatory before emergency department visits in 2024 and rapidly became overwhelmed by demand;²⁹ AI-supported triage is now being introduced to increase response capacity, beginning with acute respiratory illness.³⁰ This serves as a reminder that AI is increasingly being positioned not as a luxury, but as part of how strained health systems remain viable. Many of these projects are still in their infancy, and their real-world impact is yet to be fully established. What they share, however, is a clear direction of travel: away from novelty headlines and towards digital tools becoming sustainable, embedded infrastructure in our NHS. As the NHS 10 Year Plan signals,⁷ the question for general practice is increasingly not whether these tools arrive, but how quickly, and how safely, they are adopted.

Preparing GPNs for the integration of AI: Takeaways

For GPNs looking to begin engaging with AI in a practical and responsible way, the following are some accessible starting points:

• Track CPD and streamline NMC revalidation: Use AI to build and maintain a log of your continuing professional development learning activities, making revalidation evidence easier to compile when it’s due.
• Create clinical data templates: Ask AI to generate custom spreadsheets for monitoring patient cohorts, such as COVID vaccine eligibility or patients overdue for QOF reviews, then populate them with your own practice data.
• Advocate for AI-assisted administration: Make the case to your practice management team for trialling AI tools for admin tasks and clinical note transcription, reducing the documentation burden for GPNs and freeing up time for patient care.
• Enhance education and induction: Use AI to help design quizzes and learning activities for student nurses or new GPNs, supporting structured, engaging development and clinical education without adding significantly to your own preparation time.
• Use AI as a sounding board to brainstorm and problem-solve: When you're stuck and don't have a colleague to hand, AI can be a surprisingly useful thinking partner to bounce ideas off.
• Improve patient communications: Let AI draft social media posts, website content and batch text messages aimed at patients, saving time on routine health messaging. Always review content carefully for clinical accuracy and appropriate tone before publishing.

Conclusion

The question being asked is no longer whether AI should be used, but under what conditions it can be trusted, with confidence depending on the safety and effectiveness of tools, and on continuous human involvement and clinical sign-off.

When big changes arrive, general practice can often feel isolated from the wider NHS, and the AI rollout is at risk of repeating that pattern, with secondary care already receiving tools and support that have yet to reach our sector. My hope is that GPNs will engage with AI early and critically, understanding what these tools can offer, from easing pressure on a strained system to improving staff wellbeing, recruitment and retention. To do this, GPNs must be properly supported, whether through national training programmes or locally led 'digital champion' roles. None of this is automatic: it depends on GPNs being informed, trained, and actively involved in shaping how these tools are used. If we get that right, AI could significantly improve both staff wellbeing and the care our patients receive, while keeping what matters most exactly where it belongs: the relationship between clinician and patient.

AI use declaration

The author used Claude (Anthropic) as a research and editorial assistant during the preparation of this article. AI was used to locate and verify source material, check factual accuracy, suggest structural revisions, and provide feedback on drafts. All arguments, clinical judgements, personal examples and editorial decisions are the author's own. The author reviewed and edited all AI-assisted content and takes full responsibility for the accuracy and integrity of the published work.

References

1. KFF. KFF Tracking Poll on Health Information and Trust: use of AI for health information and advice; 2026. https://www.kff.org/public-opinion/kff-tracking-poll-on-health-information-and-trust-use-of-ai-for-health-information-and-advice/

2. Philips. Future Health Index 2026; 2026. https://www.philips.com/a-w/about/news/archive/standard/news/press/2026/philips-future-health-index-2026-ai-is-already-saving-clinicians-time-and-delivering-measurable-impact-in-healthcare.html

3. KFF. KFF Tracking Poll on Health Information and Trust: update on common vaccine myths; 2026. https://www.kff.org/public-opinion/kff-tracking-poll-on-health-information-and-trust-update-on-common-vaccine-myths/

4. Chustecki M. Benefits and risks of AI in health care: narrative review. Interact J Med Res. 2024. https://doi.org/10.2196/53616

5. Nursing & Midwifery Council. The Code; 2018. https://www.nmc.org.uk/globalassets/sitedocuments/nmc-publications/nmc-code.pdf

6. Information Commissioner's Office. Definitions; 2022. https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/artificial-intelligence/explaining-decisions-made-with-artificial-intelligence/part-1-the-basics-of-explaining-ai/definitions/

7. Department of Health and Social Care. Fit for the future: 10 year health plan for England; 2025. https://www.gov.uk/government/publications/10-year-health-plan-for-england-fit-for-the-future/fit-for-the-future-10-year-health-plan-for-england-executive-summary

8. NHS England. Artificial intelligence (AI) and machine learning; 2025. https://www.england.nhs.uk/long-read/artificial-intelligence-ai-and-machine-learning/

9. NHS England. The national context for digital primary care; 2025. https://www.england.nhs.uk/long-read/the-national-context-for-digital-primary-care/

10. Department of Health and Social Care. The future of healthcare: our vision for digital, data and technology in health and care; 2018. https://www.gov.uk/government/publications/the-future-of-healthcare-our-vision-for-digital-data-and-technology-in-health-and-care

11. Benson T. Readiness for five digital technologies in general practice: perceptions of staff in one part of southern England. BMJ Open Qual. 2022;11(2):e001865. https://bmjopenquality.bmj.com/content/11/2/e001865

12. RCGP, Nuffield Trust. How are GPs using AI? Insights from the front line; 2025. https://www.nuffieldtrust.org.uk/research/how-are-gps-using-ai-insights-from-the-front-line

13. Care Quality Commission. GP mythbuster 109: use of artificial intelligence (AI) in GP services; 2025 https://www.cqc.org.uk/guidance-providers/gps/gp-mythbusters/gp-mythbuster-109-artificial-intelligence-gp-services

14. NHS England. NHS urges 'tap the app' as 1 in 4 miss appointments; 2026. https://www.england.nhs.uk/2026/03/nhs-urges-tap-the-app-as-1-in-4-miss-appointments/

15. NHS England. NHS AI expansion to help tackle missed appointments and improve waiting times; 2024. https://www.england.nhs.uk/2024/03/nhs-ai-expansion-to-help-tackle-missed-appointments-and-improve-waiting-times/

16. NHS England. Guidance on the use of AI-enabled ambient scribing products in health and care settings; 2025. https://www.england.nhs.uk/long-read/guidance-on-the-use-of-ai-enabled-ambient-scribing-products-in-health-and-care-settings/

17. Great Ormond Street Hospital for Children NHS Foundation Trust. GOSH-led trial of AI-scribe technology shows 'transformative' benefits for patients and clinicians across London; 2025. https://www.gosh.nhs.uk/news/researchgosh-led-trial-of-ai-scribe-technology-shows-transformative-benefits-for-patients-and-clinicians-across-london/

18. Healthcare Financial Management Association. Artificial intelligence in the NHS; 2025. https://www.hfma.org.uk/publications/artificial-intelligence-nhs

19. NHS England. 500,000 NHS staff to get new artificial intelligence tools to help free up more time for patients; 2026. https://www.england.nhs.uk/2026/06/500000-nhs-staff-to-get-new-artificial-intelligence-tools-to-help-free-up-more-time-for-patients/

20. Philips. Future Health Index 2025: building trust in healthcare AI; 2025. https://www.philips.com/a-w/about/news/future-health-index/reports/2025/building-trust-in-healthcare-ai.html

21. MHRA. National Commission into the Regulation of AI in Healthcare: research, engagement and call for evidence findings; 2026. https://www.gov.uk/government/publications/national-commission-into-the-regulation-of-ai-in-healthcare-research-engagement-and-call-for-evidence-findings

22. MHRA. Guidance: medical device stand-alone software including apps (including IVDMDs); 2023. https://assets.publishing.service.gov.uk/media/64a7d22d7a4c230013bba33c/Medical_device_stand-alone_software_including_apps__including_IVDMDs_.pdf

23. TORTUS AI. TORTUS becomes the first AVT to achieve UKCA Class IIa certification; 2026. https://tortus.ai/

24. MHRA. MHRA landmark report reveals public views on AI in healthcare; 2026. https://www.gov.uk/government/news/mhra-landmark-report-reveals-public-views-on-ai-in-healthcare

25. Catto JWF, Khetrapal P, Ricciardi F, et al. Effect of robot-assisted radical cystectomy with intracorporeal urinary diversion vs open radical cystectomy on 90-day morbidity and mortality among patients with bladder cancer: a randomized clinical trial. JAMA. 2022;327(21):2092-2103.

26. Kim S, An H, Cho HW, et al. Role of artificial intelligence interpretation of colposcopic images in cervical cancer screening. Healthcare 2022;10(3):468.

27. Olawade DB, Wada OZ, Odetayo A, et al. Enhancing mental health with artificial intelligence: current trends and future prospects. J Med Surg Public Health. 2024;3:100099.

28. University College London. AI model trained on de-identified data from 57 million people; 2025. https://www.ucl.ac.uk/news/2025/may/ai-model-trained-de-identified-data-57-million-people

29. Goiana-da-Silva F, Amorim-Lopes M, Correia F, et al. The growing pains of the 2024/2025 Portugal's NHS telephone triage system national rollout. Front Public Health. 2025;13:1694713.

30. The Portugal News. SNS 24 helpline to use AI; 2025 Dec 6. https://www.theportugalnews.com/news/2025-12-06/sns-24-helpline-to-use-ai/927404