With that in mind, it has become a matter of high priority that the diverse workforce which makes up the NHS are equipped and empowered to use AI to improve patient care, whilst maintaining an ethical approach to the risks involved.

In the view of Blake and Das (2023), the implementation of AI is an impertive to relieve pressures on staff in the NHS. The question which remains is not 'if', but 'how' should be utilised.

Having located this through staff internal communications, it is indicative of the increasing awareness of how staff across hospitals are beginning to engage with Copilot, which currently lacks the essential direction on how to treat this in a literate manner. Clinical and non-clinical staff will have interaction with this tool as one of the most immediately accessible AI platforms that is available, which makes informing them about what should and should not be uploaded, as well as good information governance practice, vital.

Published in 2024, this article has already gained 111 citations, according to Web of Science, and the journal Nursing Open has a rigorous peer review process where it is a requirement that submissions should present new findings, whether good or bad. One of the authors, Khan Rony, has relevant positions in the US and Bangladesh.

‌

Supplied by the recognised professional body for doctors and medics, this document includes its own principles of AI usage, which can be disseminated within hospital settings.

The article makes a strong case for educating the wider public to have better awareness and understanding of how machine learning is safe and meets regulatory standards. Published by the World Health Organization, it meets expectations as a source that is broadly trustworthy.

‌

Despite concerns noted on ethics and individual capability, it does cautiously advocate for AI to be integrated as part of the work practice and skills which are important for community nursing. Published in the British Journal of Community Nursing, hosted as part of the Mark Allen Group, it is part of the reputed journal which is dedicated to this particular profession, and all content is peer reviewed.

‌

There is a much broader picture detailed too, which contextualises the role of procurement of AI systems might affect patient experience, with one of the significant risks highlighting that siloed procurement and concentrating AI in some areas can still result in overwhelm elsewhere. Overall, it is an effective analysis which outlines the practical usage of AI in services while providing a high-level insight on what needs to be done centrally.

There is an inclusion of a generated response from Copilot, which seems to corroborate the view that it can create efficiencies in written documents and allow them to concentrate on patient care, but not to act as substitutes. It is important to note this piece has a distinct position, but provides a critical voice that is likely to be reflected in the workforce.

‌

While researchers might be the target audience that work within healthcare based institutions, this will be relevant to library professionals and students who will be conducting literature reviews.

‌

There are issues which need to be interpreted in the context of best practice in healthcare, where AI tools for decision making can be applied when human input is ‘meaningful’, it is typically advised against with patient care.

‌

UK Government (2025) Fit for the Future - 10 Year Health Plan for England. UK Government.

‌

Hopkins, E., Smith, S. and Wood, H. (2024) ‘Adoption and everyday use of artificial intelligence by NHS knowledge and library professionals in England’, Journal of EAHIL, 20(2), pp. 11–15. Available at: https://doi.org/10.32384/jeahil20619.

‌

S.C. Shelmerdine et al. (2024) ‘Artificial Intelligence (AI) Implementation within the NHS: The South West London AI Working Group Experience’, Clinical Radiology, 79(9), pp. 665–672. Available at: https://doi.org/10.1016/j.crad.2024.05.018.

‌

‌

Incorporating health literacy practice, the article advocates for patients to be effective partners in shaping AI tools to help overcome personal reservations about how their data is used. Both Thompson and Palmer equally alluded to examples that without transparency, data consent will unlikely be approved.

‌