Predicting the future: is the eternal fantasy about to come true?


Predicting the future is on the verge of becoming more reliable than ever: science has replaced magic, and algorithms that structure the data available on the Web have replaced prophets and wizards. The prediction market is likely to skyrocket soon, perhaps making a handful of people very rich. But this new technology will also find applications in global aid and solidarity


Predicting the future has always been one of humanity’s fantasies. Who has never dreamed of getting a glimpse of what their life will be like ten years from now? Who never wished they knew, when on the verge of making a difficult decision, what the outcome was going to be? Down through the ages, in every culture, this urge to know has paved the way for billions of augurs, witches and gurus to warn, reassure and fleece the tormented souls who fear the future. Today, we (collectively) may not be as superstitious as we used to be, and we may not believe in witches anymore, but a new kind of magic has been making its way into our everyday lives: science. Science, and more particularly computer and Internet science, reactivates the millennia-old fantasy of forecasting the future.

Only today, it is generally no longer seen as a way to find out our own personal prospects, but rather to predict and prevent global events in order to curb their potentially harmful effects.

Ever since we have been able to see our planet from above, we have realised that we all embarked on the same ship, and that individual destinies are intrinsically linked with the planet’s fate. This lays the ground for a new quest for prediction. Fear is still a part of the equation, though.


Sanitary prediction is the next big thing


In 1942, in A Study of War, American political scientist Quincy Wright wondered whether conflicts could be predicted, based on public opinion. Ever since, the intelligence community has made it its business to foresee the future, to predict potential wars, revolutions, outbursts of violence or new demands from the people. International relations experts have long learned to analyse the tiniest hints of change that could be indices of upcoming brutal shifts. The irruption of Internet in the world of communication has dramatically widened the possibilities: the immense mass of intel produced every day on the Web is an endless source of knowledge – and speculation. Simultaneously, predicting the future ceased to be the prerogative of the intelligence community, and upcoming wars ceased to be the main concern. In 2006, Google launched Google Flu Trends: based on the occurrence and pertinence of certain search terms (there is apparently a “close relationship between how many people search for flu-related topics and how many people actually have flu symptoms”) this service monitors and estimates the evolution of flu activity. A similar service has been issued to track dengue activity. In fact, sanitary surveillance is the next big thing in the world of predictions. Just a few weeks ago, researchers from Microsoft and the Technion-Israel Institute of Technology announced they had built software that could predict upcoming catastrophes (mainly extreme climatic events) and the diseases that were likely to hit in their aftermath. This software also relies on the information available on the Internet, thanks to 22 years of archives from The New York Times, and to tools like DBpedia, a crowd-sourced community effort that structures Wikipedia data into a workable database, and WordNet, which helps the software understand the meaning of words. In total, the system draws on around 90 different sources. In its testing phase, it retroactively predicted a cholera outbreak that happened in Angola in 2006 after a period of droughts, « because previous events had taught the system that cholera outbreaks were more likely in years following droughts,” wrote the MIT Technology Review in an article about the project.

In similar tests involving forecasts of disease, violence, and a significant numbers of deaths, the system’s warnings were correct between 70 to 90 per cent of the time.

These results make us wonder: is humanity’s old dream about to come true? Will we one day be able to predict virtually anything with 100% accuracy?In any case, it seems that science, thanks to the enormous amount of data that mankind has accumulated online, is doing much a better job than all the psychics there ever did. But the real question, for now, is: to what end? How will we use our knowledge of the future?


Good news for the business… and for aid organisations


Microsoft and Technion have already announced that they do not plan to commercialise their system yet, but research continues. Eric Horvitz, co-director at Microsoft Research, told the MIT Technology Review they wanted to integrate even more newspaper archives and digitized books into the database. To him, the software has already proved efficient enough that concrete applications can be found for it in real settings,

to assist experts at, for example, government aid agencies involved in planning humanitarian response and readiness,

says the MIT’s publication. Aid organisations and NGOs could in fact find this tool extremely useful in tackling disease outbreaks. Knowledge of the future would then become a synonym of being empowered to protect populations from the catastrophes that could possibly strike them. Knowledge is power, and with great power comes great responsibility. Maybe that is what is now encouraging Microsoft and Technion to say they want to use their work to serve mankind, and in particular the most exposed populations. Cynics might say that it is not going to take long before everyone fights to make money out of such a powerful tool. In fact, one could become immensely rich, “just” by being able to predict the future of macro-economic cycles, for instance. But there are already other people, with their own technology, on the starting-blocks for that kind of use of prediction. Swedish-American start-up Recorded Future, for instance, also continuously scans high-quality material available on the Web to allow its clients to “explore the past, present and predicted future of almost anything in a matter of seconds.” Several hedge funds already use Recorded Future, whose strapline is “Unlock the predictive power of the web,” to determine their investment choices. The start-up has about 100 subscribers, and can count on the support of two very important funders: In-Q-Tel, the CIA’s investment arm, and Google Ventures, writes NPR (National Public Radio, USA). This is indication enough that the prediction market is likely to soar in years to come, as the technology becomes more refined and accurate. And in fact, its becoming a major business development lever might be good news for the world of philanthropy and aid organisations: when predicting the future has become easy and banal, it will also be more accessible to all. While we do need, for now, initiatives that concentrate their efforts on exposed populations for free, predicting and preventing catastrophes will hopefully one day also be the prerogative of governments, security agencies, development agencies, NGOs, etc.

Danish physicist Niels Bohr (1885-1962) once said: “Prediction is very difficult, especially about the future.” We may think we are about to counter his assertion, but it should also come as a healthy reminder: no matter how much science we pour into our everyday lives, there is something about both humanity and nature that is implacably unpredictable. And that also makes for the magic of life. Good news: the fear of the future might not be completely eradicated, after all.

- By Usbek & Rica -

Photo © Shutterstock / Astrobobo