A former Swedish Army ranger with a Ph.D. in computer science, Alberg is chief executive of Recorded Future, a Cambridge, Mass.-based firm that has pioneered real-time use of the Web and social networks as a way to predict events in the near future. Recorded Future's computers continually scour tens of thousands of Web sites, blogs and Twitter accounts, and use sophisticated analytical software in an effort to spot "invisible links" between items that actually refer to the same people and future events in which they may be involved. Recorded Future also tracks the volume of such links, and then tries to use that information to gauge the momentum behind a future event, the likelihood that it will happen, and when and where it may occur [source: Shachtman].
Recorded Future's technology has enough potential that both Google and the U.S. Central Intelligence Agency have become investors [source: Shachtman]. But as Alberg admitted in a 2011 interview with Business Insider, the software has limitations, in that it's better at some sorts of predictions than others. It does pretty well with forecasting events that frequently occur, such as stock market volatility, but not so well with infrequent events, such as elections. And predicting so-called "black swan events" -- random, seemingly unlikely disasters, such as the cascading near-collapse of Wall Street in 2008, that somehow happen despite the odds against them -- remains a daunting challenge [source: Rosoff].