Introduction:

Artificial Intelligence (AI) has begun to play a transformative role in healthcare, promising to reshape the industry with improved outcomes and efficiency. The Leverhulme Centre for the Future of Intelligence’s launch event highlighted AI’s potential to be the most significant event in our civilisation, a sentiment echoed by Stephen Hawking. This article examines AI’s burgeoning impact on healthcare and addresses the balance between its benefits and risks.

The Rise of AI in Clinical Applications:

AI’s integration into healthcare is rapidly advancing. It is employed in disease detection, real-time outcome prediction, treatment personalisation, and administrative task streamlining. AI’s capacity to augment human decision-making in these areas is remarkable. For instance, AI models have significantly reduced the drug discovery timeline, a boon for pharmaceutical development.

Navigating the Risks of AI in Healthcare:

Despite the promise of AI, risks such as bias, privacy breaches, and workforce displacement loom large. The datasets and models AI relies on are human creations, subject to human error and prejudice. Recognising and addressing these risks is crucial to prevent exacerbating healthcare inequities and introducing new ones.

Frameworks for Responsible AI Use:

A multitude of frameworks have emerged to guide the responsible use of AI in healthcare. The Asilomar AI Principles, for example, set forth guidelines for AI development and usage. However, the rapid evolution of AI technologies, such as large language models, presents challenges to the practical application of these frameworks, highlighting the need for adaptive governance.

Towards a Harmonised AI Code of Conduct:

A proposed AI Code of Conduct aims to harmonise principles and commitments for stakeholders across the AI lifecycle. This framework builds on the core principles of a continuously learning health system (LHS) and reflects a consensus-driven approach to fostering responsible AI use in healthcare.

Conclusion:

AI offers an unprecedented opportunity to realise the vision of a learning health system. However, its application must be managed collectively and responsibly. The proposed AI Code of Conduct provides a foundation for stakeholder engagement and the co-creation of a future where AI-enabled health, healthcare, and biomedical science serve the well-being of all.