The superhero movie genre has been around for almost a century, but recent popularity can be traced to Superman (1978) with the late Christopher Reeve, with a notable change in 1989 with Tim Burton’s Batman, when the attention to special effects and atmosphere created the visual equivalent of the Comic Book vibe.  Now the latest trend is teams of superheroes battling some archetypal Jungian menace (Avengers:  Age of Ultron) or (now that all the bad guys have been whipped, I guess) teaming up and fighting each other out of sheer adolescent boredom (Captain America: Civil War). 

I attended an excellent Data Driven Meetup event recently in NYC (ably curated and sponsored by FirstMark Capital and Matt Turck) where one of the presenters was Peter Brodsky of HyperScience (intriguingly, their web site says “We’re currently in stealth”).  He presented what appear to be significant advancements in capabilities around breaking ML tasks down into subnets in order to create re-usable building blocks which take less time and less data to train, and which can be “cross-trained” for improved accuracy.  His presentation created tremendous positive buzz and optimism about where we all are going with AI/ML.  These days, when the somewhat shy and deep AI/ML community gathers, in addition to there usually being an understated positive atmosphere, you also often hear darkly humorous references- “well this is all good until we all lose our jobs to therobots.”  But is that undercurrent really a worry about innovation, or is it more a concern about where the world is going in general?  Are the superheroes winning or losing?  How is this movie going to end?  What a cliffhanger. 

In the heady stew of the global pop culture, dark thoughts in the middle of the night about dystopian futures, and the current work on orchestrating Data, AI, and ML (D-AI/ML), we need to wake up every day and decide what to do next.  We work with data to help our companies, and sometimes hope in our own small way to be superheroes who can travel back in time (linear regression analysis), see the future (risk analytics), see through walls (next best offer analysis) and bend steel with our bare hands (commodity futures analysis).  We may not be saving the world right at this moment from some dark menace but we might be able to save our employer from writing a too-inexpensive life insurance policy to a skydiver.  It’s a start.  

There are some themes which come to mind for me: 

  • We need to work as a team with different “superpowers” to orchestrate innovation and cooperation-- open source, open shareable data (D), artificial intelligence (AI), machine learning (ML), and many varied human skillsets
  • We need to strive to understand the combinatorial design patternsfor D, AI, and ML that work and bring them quickly to effective real-world fruition as building blocks of a better enterprise and a better world
  • The immediate and real evil forces in business and life are not about robots taking over the world and not as elegantly sinister as the bad guys in the movies-- and there are clearly applications of Data, AI, and ML that can help
  • We need to support the directed use of these capabilities for the greater good, each in our own way 

If we focus on these principles, maybe we can live up to that imaginary cape.

 Elevondata (www.elevondata.com) is a leading edge data management advisory and data lake solutions company.  Vin Siegfried, who does not own a cape but does own a Superman T-Shirt, is one of the founders and can be reached at vsiegfried@elevondata.com.  Vin is also the Chief Operating and Technology Officer for AtrocityWatch (www.atrocitywatch.org) a not-for-profit dedicated to using technology for humanitarian goals.

Author Vin Siegfried