The world’s data is expected to double every two years for the foreseeable future, thanks to increasing connectivity and progress in technology. This abundance of data combined with advances in computing power, software and algorithms have allowed us to do tremendous tasks such as push the boundaries of artificial intelligence that helps us speak with virtual assistants on our phones or get better product recommendations on websites.
However, as the digital world yields more and more data, questions surrounding its responsible and equitable use are also beginning to rise. Aside from issues of responsible use that deals with privacy and ethics, issues of equitable use are also becoming pertinent – especially on who gets to control and utilize the vast troves of data out there. With recent instances of mismanagement around personal data, the development of stricter regulations and oversight, and the pursuit of big data as a competitive advantage by large companies, data is increasingly being siloed as organizations focus more on building boundaries around their data assets. This lends credence to the claim that even though much of the world’s data was generated only recently, only a tiny fraction of it (some say just 1 percent) has ever been analyzed. To put it simply, the world is currently not geared towards collaborative value generation around big data.
A good indication on what the future holds for collaborative use around big data is in the European Union’s (EU) Communication on Building a European Data Economy. Here, the EU defines the ‘data economy’ as, “an ecosystem of different types of market players – such as manufacturers, researchers and infrastructure providers – collaborating to ensure that data is accessible and usable. This enables the market players to extract value from this data, by creating a variety of applications with a great potential to improve daily life (e.g. traffic management, optimization of harvests or remote health care).” Accordingly, along with research institutions and government agencies, the private sector is also developing data collaborative arrangements where organizations are enabling the responsible sharing of their data assets because they realize that these assets can generate further public value elsewhere.
These are encouraging developments, because it is only when we open-up data and begin treating it as a public good that individuals and organizations can start innovating around it. As a result, it is reasonable for us to predict that the next leg of use-cases around big data will be realized through the collaborative use of data that is responsible and ethically sound. At the Digital Impact Alliance (DIAL) we believe that such collaborative and responsible use has the potential to unlock significant impact within the humanitarian and non-profit sectors too.
We are already seeing encouraging instances of collaborative value generation around big data in the non-profit sector. For example the GSMA, a trade body that represents the interests of mobile operators worldwide, recently demonstrated how mobile data can be used to mitigate the impact of natural disasters, infectious diseases and pollution. The extent to which these data can be responsibly accessed and utilized will determine the broader socio-economic outcomes in our digital future. Aside from mobile data, third party data such as geospatial data is being operationalized to accurately map the population in remote and rural areas that could aid the humanitarian sector. In the field of health research, we are seeing data commons that allow researchers to pool together data assets and collaboratively work toward a single shared goal. These data commons are set up to enable access to the data in a way that protects sensitive information – which is critical to managing concerns of privacy around health data.
Attempts are also being made at decentralized approaches of connecting global data marketplaces to incentivize the sharing of data for research, commerce and development purposes. To complement such opening up of data, encouraging instances of techniques such as federated learning are also being developed that could train machine learning algorithms without compromising privacy.
Here at DIAL, we have collaborated with mobile network operators and government agencies in Malawi to build a model to forecast migration patterns and optimize resource planning in order to maximize access to health facilities by the population. Additionally, our partners Flowminder and OPAL are building open source software that enable access to and analysis of anonymized mobile data at national scale. As regulation and norms around responsible use mature, we will continue to actively contribute to the discussion. We will also ensure that DIAL’s work is informed and continues to adapt to relevant data protection guidance and considerations as well as the Principles for Digital Development which DIAL stewards.
As the world moves towards more collaborative use of data, DIAL remains committed to having open conversations, participating in constructive feedback and sharing our learnings so we can collectively drive towards achieving the Sustainable Development Goals. We must continue to push the ecosystem past the current paradigm of ad-hoc partnerships and lack of transparency in the use of data; and seek ways to collaboratively unlock the true potential big data has to offer.
Hitoishi Chakma works as a Fellow for the Data for Development team at the Digital Impact Alliance.