Nick Bostrom, Swedish philosopher and head of the Future of Humanity Institute (FHI), has a definite concern on the future of AI. Though he promotes and seeks advancement in AI technology, he believes that AI is the present nuclear threat of our century. Within the last 5 years alone, “…three of AI’s ‘Holy Grails’ have been captured: recognizing objects in images, speech recognition, and machine translation” (Stuart Russel, British AI expert). But there is more “Holy Grails” to be captured. For example, teaching AI to understand and read language. The question is who will be next to have a breakthrough in AIs technology and what will they use technology for?
Bostrom believes that with embryo selection on the horizon, that those to make key breakthroughs in AI will nonetheless be genetically enhanced scientists. This may seem far-fetched and decades away but, in all honesty, embryo selection is attainable within the next 5 years. So, will these genetically enhanced scientists use AI for the transhumanist movement, a movement which aims and seeks to extend human lifespans? Ultimately, creating a world utopia? A future Nick Bostrom hopes to be a pioneer for. Or, will these breakthroughs be used to bring our pending doom?
In one of his latest papers, “The Vulnerable World Hypothesis”, Bostrom discusses the fear he has of “…those tinkering with ‘black ball’ technologies”. Black ball technologies including that of AI and proposes “global governance” as a solution to keep human enhancement ethics in check. However, this would include the government spying on our every move through infringement of our personal privacy. Arguably, this is already being done with conspiracy theories suggesting that the government is listening to our every conversation and move through our social media accounts, internet use, cellular devices and the smart technologies that are now used in our homes. Regardless, Bostrom believes that global governance will aid in alerting authorities to those “…tinkering with ‘black ball’ technologies” and keep the holy grails of AI from the wrong hands. But is global governance really the only solution? Bostrom dauntingly states that we are, “…in the middle of doom and utopia.” But can we rely on global governance to keep us from doom and put us on track for future utopia?
Quetteville, H. (2019, January 6). Will AI be this 21st-century’s nuclear threat? The Telegraph. Retrieved from https://www.telegraph.co.uk/technology/0/will-ai-21st-century-nuclear-threat/