Big data applications are increasingly shaping our everyday lives, making trends more transparent and patterns more predictable. Like insurance, science, notably medicine, is equally subject to the novel possibilities and demands resulting from big data. Will risks gradually decline as trends and patterns become more predictable in healthcare and for behaviour? Are the days of insurance and science as we know them numbered?

Insurers and scientists thrive on uncertainty. The inability to determine the future has left its mark on how the scientific community interprets empirical knowledge claims (theories), i.e. as being in principle subject to replacement by better ones. To put it bluntly, theories are uncertain. The eighteenth century French mathematician, Pierre-Simon Laplace, like many others in his day, was convinced that uncertain or probabilistic knowledge was not inherent in our world but resulted from insufficient information (read: data).

Following Laplace's thinking, if a superior intelligence knew the totality of natural laws and present physical states, it would be able to infer the future course of events with certainty. The need for theory would end.

Some experts believe that big data will allow us to reach the unprecedented possibility of approaching such Laplacian intelligence. A simple example supports their claim: when we are unable to observe every swan on earth at any given point in time, the assertion that “All swans are white” can only be inferred from a limited number of observations and must therefore remain fallible. We cannot rule out refuting it upon the first observation of a black swan. If, however, we are able to observe all swans on earth at any given point in time and find them to be all white, then “All swans are white” will no longer be a theory but will have become a matter-of-fact statement, such as “The sun is shining today”.

Adding to this supposed trend of diminishing risk is the accelerating pace of social change, which is itself exacerbated by big data. Social change has brought about increasing fragmentation in how people plan and live. Previous forms of stability and certainty no longer apply. Therefore the ultimate target of insurance, the status quo, will be changing in ever shorter intervals. Insurers now need to question whether people will want long-term products and use them, as they have in the past.

At the same time, big data and social change provide information on challenges and opportunities for life insurers who need to respond to profoundly changing client needs in addition to new ways of assessing risk, notably the potential replacement of medical data by socioeconomic and behavioural data as predictors of insured events.

As consumers increasingly opt for short-term covers that match their lives, the need for long-term guarantees, one of the major concerns for life insurers at present, will likely diminish.

For more on big data, read the articles I wrote with my colleague Thomas Gehling: “Big Data, Big Insight – Is Knowledge Still Power in a Digital World?” and “Big Data, Big Insight – What Does it Offer to Life Insurers?”