Technology touches every facet of our lives. As we go from day-to-day connected to the ‘internet of things’ most of us are oblivious to the vast quantity of big data we’re generating as a by-product of interacting with our electronic devices. The global rise in the adoption of technology coupled with unprecedented computing power has engendered a worldwide data revolution where artificial intelligence, machine learning, semantics and data mining are key to corporate, governmental and political success.
As ‘The Economist’ reports that ‘the world’s most valuable resource is no longer oil, but data’, it’s true to say that mining the ‘oil of the digital era’ is lucrative for some but raises ethical, socioeconomic and political concerns for many.
Our modern system of internet mass surveillance suggests that George Orwell’s predictions in his novel, ‘Nineteen Eighty-Four’, may not have been too misguided after all. The growing capability for transforming our online existence into data points for exploitation raises serious privacy concerns. Most citizens are unaware of how, when and what data is collected about them and with whom it is shared. A recent study by Oxford University (Binns et al., 2018) reveals that tech monopolies, like Alphabet (Google’s parent company) and Facebook, hoover-up and monetize an enormity of our data from third-party apps on our smartphones.
What’s more, machine modelling of citizen data can lead to ‘social sorting’ whereby individuals are categorised on the basis of their digital data identity. Algorithms are not always certain but can be biased resulting in a void between virtual information and the physical truth which can translate into embedded injustice. According to the Data Harms Record (), automated data-led decisions can drive discrimination influencing an individual’s ability to access housing, insurance, education or a job, thereby reinforcing long-term socioeconomic inequality and perhaps further exposing those individuals to poverty.
Despite the harms, there are many real-world examples of how big data can be used for wider societal benefit.
The GSMA’s ‘Big Data for Social Good’ project, launched in 2017, has established a holistic approach to using citizen data captured on mobile telephone operator networks for mitigating various humanitarian crises. The first round of trials showcase big data capabilities; Bharti Airtel and Be He@lthy, Be Mobile (a joint WHO and ITU initiative) are working together to control the tuberculosis (TB) endemic in India. Mobile data provides current behavioural information on a population of around 280 million people which can be used to identify TB hotspots and help target public health interventions, such as vaccines. Similarly, Telefόnica are using mobile data in São Paulo, Brazil, to improve city traffic management and environmental planning with the broad aim of reducing air pollution.
Non-profit organisation, DataKind, is another institution that’s aims to harness the power of citizen data in the service of humanity. Their mission is to collaborate with high-impact organisations and social enterprises, leading data scientists, developers and designers on diverse projects to maximise social impact. Examples of recent projects include the use of data to create paths out of homelessness, improving college success through predictive modelling, and using satellite data to find villages in need. Earlier this year, DataKind announced raising $20M investment to ‘Support the Data Science for Social Good Ecosystem’ meaning that they will transition from a project to a platform based model capable of supporting more organisations.
Amongst those enterprises promoting the use of data for social good are those which are advocating responsible data practice to minimise harms. Aside the European Union’s General Data Protection Requirements (GDPR) that were enforced in May 2018 to stipulate rules of data capture, storage, usage and sharing, a number of organisations have focused on the regulation of data sharing and management in order to create a safe and trustworthy open data ecosystem. For example, The Open Data Institute (ODI) is working with central and local government on the UK’s first ‘data trust’ pilots. Although somewhat ambiguous, in this context the term ‘data trust’ refers to a framework or legal structure that provides independent third-party stewardship of data in the aim of safeguarding data and building trust whilst increasing access.
Other initiatives, such as the BYTE Big Data Community (BBDC), provide guidance for industry and other big data practitioners on how to integrate the social concerns with emerging big data management practices. Likewise, the ERC-funded DATAJUSTICE research project is seeking effective ways of ‘approaching data justice in the age of datafication’.
In a world where data rules, it is important that we continue to raise awareness of the ethical, socioeconomic, and political issues associated with big data dissemination. It is important to stimulate debate and critical interrogation of how data use and availability should be regulated today and in the future. Collaboration and joined-up thinking across multi-disciplinary sectors can help in achieving data justice for all; this is the only way we can ensure technological innovation can proceed unhindered.