RSS executive director Hetan Shah explains why the Royal Statistical Society is pushing for a Council for Data Ethics, which could develop new frameworks for data governance and rebalance the current 'data trust deficit'
We live in a fast-changing landscape of digital data. Algorithms, big data, data science, machine learning, the internet of things and smart cities—to cite just the most recent trends—are changing policy, business and daily life.
Every day brings news stories reporting the ethical dilemmas that this data innovation is creating. Old notions such as privacy and consent, which have long governed ethics around data, are being stretched to breaking point. New questions have emerged, such as how we can hold algorithms to account or who is at fault when a driverless car crashes.
Last year, the Royal Statistical Society made the case in its Data Manifesto that we must harness data for the good of society, but in a way that maintains public trust. Such trust is not a given—a survey by Ipsos MORI conducted for the RSS in 2014 suggests that there is a general data trust deficit: people have less trust that an institution will use their data appropriately than they do of that body in general.
Public support for sharing personal data depends very much on who it is being shared with, and why. The evidence shows that when a case for public benefit is clearly stated and when there are safeguards in place, more of the public take a positive view of data use and sharing than disagree due to privacy risks. The addition of safeguards, such as anonymisation of data, or punishment for data misuse, significantly improves support from 33% to around 51%.
One message for policymakers therefore is that they need to clearly communicate the value of any data sharing initiatives, and they need to put safeguards in place. It is also noteworthy that there is considerable opposition to sharing data for commercial purposes, making this is an area where policymakers must tread very carefully.
We should remember, however, that what underpins trust is trustworthiness. Therefore government must consider how it can ensure that its usage of personal data is trustworthy. This means ensuring security, privacy, good governance and equitable treatment. These has been lacking in some high-profile initiatives, such as the NHS’s abortive care.data scheme.
Data innovation will not wait for us. If we do not stop to think about these issues and address them proactively, public trust could rightly be lost. This could seriously set back innovation for public good, repeating the pattern seen previously for genetically modified crops and more recently with the sharing of health records.
The RSS has argued, therefore, that the time is right for a new independent institution to step back and consider the ethical implications of evolving data technologies. Such a body—a Council for Data Ethics—would have the space to develop new frameworks and norms that in turn lead to institutions and data governance in the public interest.
An ethics council would have its work cut out. Besides obvious issues such as privacy and security, there is a gaggle of other questions to work through. What if algorithms lead to racist or sexist outcomes? Or what if they are less discriminatory than solely human judgements? What governance mechanisms do we need around facial and voice recognition technologies? How can we hold an algorithm to account? If you struggle to change your credit score, imagine life once algorithms are endemic in health and criminal justice.
One area where government can move ahead quickly without breaching public acceptability is in improving researcher access to data. Our polling indicated that public support was strongest, with 80% either supportive or neutral, for sharing anonymised data with university researchers conducting publicly funded research.
Researchers, however, know that getting access to government data can be very difficult. The RSS has pushed hard in the last year for the legislative framework in this area to be improved, and the new Digital Economy Bill, which is on the cusp of becoming law, creates an enabling framework for government departments to share data with researchers—no more hiding behind the excuse of data protection.
This article is based on a presentation Hetan gave to the Reform think tank’s Big Data conference and was originally published in Research Fortnight.