The Exponentially Accelerating Progress in Artificial Intelligence Raises Safety Questions

A day does not go by without a news article reporting some amazing breakthrough in artificial intelligence. In fact progress in AI has been so steady some futurologists, such as Ray Kurzweil, are able to project current trends into the future and anticipate what the headlines of tomorrow will bring us. Let’s look at some relatively recent headlines:

1997 Deep Blue became the first machine to win a chess match against a reigning world champion (perhaps due to a bug).

2004 DARPA sponsors a driverless car grand challenge. Technology developed by the participants eventually allows Google to develop a driverless automobile and modify existing transportation laws.

2005 Honda's ASIMO humanoid robot is able to walk as fast as a human, delivering trays to customers in a restaurant setting. The same technology is now used in military soldier robots.

2007 Checkers, the computer learns to play a perfect game in the process opening the door for algorithms capable of searching vast databases of compressed information.  

2011 IBM’s Watson wins Jeopardy against top human champions. It is currently training to provide medical advice to doctors and is capable of mastering any domain of knowledge.

2012 Google releases its Knowledge Graph, a semantic search knowledge base, widely believed to be the first step to true artificial intelligence.

2013 Facebook releases Graph Search, a semantic search engine with intimate knowledge about over one billion of Facebook’s users, essentially making it impossible for us to hide anything from the intelligent algorithms.

2013 BRAIN initiative aimed at reverse engineering the human brain is funded by 3 Billion US dollars by the White House and follows an earlier Billion Euro European initiative to accomplish the same. “It just so happens that the same technology the project will develop … could also be used to make our brains do whatever they want. Wirelessly. From a distance.”

From the above examples, it is easy to see that not only is progress in AI taking place, it is actually accelerating as the technology feeds on itself. While the intent behind the research is usually good, any developed technology could be used for good or evil purposes.

From observing exponential progress in technology Ray Kurzweil was able to make hundreds of detailed predictions for the near and distant future. As early as 1990 he anticipated that among other things we will see between 2010 and 2020 are:

• Eyeglasses that beam images onto the users' retinas to produce virtual reality developed. (See Project Glass)
• Computers featuring "virtual assistant" programs that can help the user with various daily tasks. (see Siri)
• Cell phones built into clothing and able to project sounds directly into the ears of their users. (See E-textiles)

But his projections for a somewhat distant future are truly breathtaking and scary. Kurzweil anticipates that by the year:

2029 Computers will routinely pass the Turing Test, a measure of how well a machine can pretend to be a human.

2045 The technological singularity occurs as machines surpass people as the smartest life forms and the dominant specie on the planet, and perhaps universe.

If Kurzweil is correct about these long term predictions, as he was correct so many times in the past, it would raise new and sinister issues related to our future in the age of intelligent machines.

Will we survive technological singularity or are we going to see a Terminator like scenario play out? How dangerous are the superintelligent machines going to be? Can we control them? What are the ethical implications of AI research we are conducting today? We may not be able to predict the answers to those questions, but one thing is for sure - AI will change everything and impact everyone. It is the most revolutionary and most interesting discovery we will ever make. It is also potentially the most dangerous as governments, corporations and mad scientists compete to unleash it on the world without much testing or public debate. I am excited to devote my next book to looking for answers to the fundamental questions raised by exponential developments in artificial intelligence and in particular its safety implications.

Dr. Roman Yampolskiy is a computer scientist and a director of Cyber Security Lab at the University of Louisville. His recent research focuses on technological singularity. Dr. Yampolskiy is currently working on a book about the safety implications of the coming technological singularity - “Artificial Superintelligence: a Futuristic Approach.”