Artificial intelligence and big data have the potential to revolutionize and democratize healthcare across societies. However, their use carries significant ethical implications. We must ensure that the right frameworks are in place so that new technologies help rather than harm, human welfare, and that we avoid missing opportunities to improve living standards.
When it comes to technology, we have already seen successful applications of AI and big data in healthcare, including data analytics in clinical trials, prevention of re-admissions, faster diagnosis of rare diseases, and better predictions of diseases like cancer. So there is an urgent need to make progress on data ethics as well. Whilst these applications may entail specific regulatory and compliance requirements, compliance is not the same as ethics. Compliance may be compared to playing by the rules of the game. Ethics is how well you play. It is therefore an ongoing process aimed at improving the game itself, which means, in this case, doing anything we can to improve human welfare whilst fostering individual and social rights.
There is a significant difference between what can be done (feasibility), may be done (legality) and should be done (ethics). Striking the balance between these three elements can be challenging, especially when it comes to the ethics of health data, a particularly sensitive kind of personal information. That is why it is vital to speed up research into the topic and invest in foresight analysis. In our digital society, the implications of technology are no longer purely legal or technical. They have become conceptual, forcing us to reconsider fundamental philosophical questions related to personhood, identity, the human condition, and the values we wish to promote.
To this end, the Digital Ethics Lab of the Oxford Internet Institute (OII) at the University of Oxford, the Data Ethics Group at the Alan Turing Institute (ATI), and Microsoft are collaborating on a new project on “The Ethics of Medical Data & Advanced Analytics”. This aims to foster research around the ethics of health data in Europe. Over the past few months, we have created a network of experts from across the fields of ethics, law, healthcare, machine learning, and industry, and brought them together to scope the ethical considerations related to adopting advanced analytics tools and exploiting health data. Over the next two years, we plan to establish best practices for the ethical use of health data, and to identify and help mitigate any risk of unethical consequences.
Convening the relevant stakeholders to tackle these issues is essential if we are to unlock the vast potential of data, which is fast becoming a most valuable resource. Within the OII’s newly established Digital Ethics Lab, we have an opportunity to collaborate on new projects such as the one supported by Microsoft on developing a European ethical code for data donation and encouraging data philanthropy, which is building on the joint “Ethics of Medical Data & Advanced Analytics” project mentioned above. The goal is to explore ways in which citizen participation in research efforts may be supported via ‘data donations’, and to shape best practice with regards to respecting individuals’ rights as well as ensuring proper regulatory oversight of existing and future data exchange partnerships between governments and tech companies.
This is an ambitious plan and it is clear that we are only just embarking on what will be a long journey. Yet it is a necessary one. Existing ethical frameworks are insufficient to address the specific challenges faced in medical data analytics, and to ensure that Europe maximises its benefits while minimising its risks. There will undoubtedly be difficulties on the road along the way but the ultimate destination – a digital world underpinned by a pro-ethical infrastructure – is well worth the effort and work.