Artificial Intelligence: Friend or Foe?


Thought Leadership

Artificial Intelligence: Friend or Foe?

Neil Farquharson

A few days ago, the Future of Life Institute (FLI) published an open letter here, signed by a large number of eminent professors and industry leaders. The impetus for the letter can be traced back to early December when Professor Stephen Hawking, the renowned theoretical physicist from Cambridge University, warned of the dangers of letting artificial intelligence (AI) develop itself. That is, putting AI in charge of designing and building newer AI.

Image via Warner Brothers


Up until now the idea of AI has largely been confined to science fiction: Isaac Asimov’s I,Robot series, recently made into a movie starring Will Smith, or the Terminator movies for example. Back in 1968 Stanley Kubrick and Arthur C. Clarke scared movie-goers half to death with 2001: A Space Odyssey when the computer HAL “had an acute emotional crisis” and started killing the astronauts. These dystopic futures have been regarded as good horror movie fare and not to be taken seriously – until now.

We’re already seeing early versions of AI being used, at least experimentally. For example in the case of driverless cars, cars that will be tested on the open highway with artificial intelligence doing the driving. Doctor Ben Metlock, one of the signatories to the FLI letter, states: "Traditionally we have a legal system that deals with a situation where cars have human agents. When we have driverless cars we have autonomous agents... You can imagine a scenario when a driverless car has to decide whether to protect the life of someone inside the car or someone outside." In other words, AI may soon be making life and death decisions autonomously.

Happily for us, automation - the little brother of AI - is still our friend. Automation can make decisions for us in real time, removing the hassle or drudgery of many decisions that would otherwise bog down our lives. One such automated system from Zix is policy based email encryption. Humans like you and me can send and receive sensitive emails in the normal manner, thereby working at our full productivity. Operating in the background, the ZixGateway detects which emails need to be protected with encryption and which do not; and the ZixGateway will make these decisions in real-time, consistently getting these decisions right.

I’m not sure that I want AI making life and death decisions for me, but I do like the ZixGateway for protecting the sensitive information in my emails.