Rise of the Machines Review: AI's broader context
Since 2012, there has been increasing attention for machine learning and especially deep learning, the ML branch that benefits from more advanced techniques and increased computing resources to equal human-level performance in some domains.<iframe style="width:120px;height:240px;float:right;margin-bottom:20px;margin-left:20px;" marginwidth="0" marginheight="0" scrolling="no" frameborder="0" src="//ws-na.amazon-adsystem.com/widgets/q?ServiceVersion=20070822&OneJS=1&Operation=GetAdHtml&MarketPlace=US&source=ac&ref=qf_sp_asin_til&ad_type=product_link&tracking_id=webn3rd-20&marketplace=amazon®ion=US&placement=1925228649&asins=1925228649&linkId=2e40136a2ca4da0b883932007f7e85d3&show_border=false&link_opens_in_new_window=true&price_color=333333&title_color=0066c0&bg_color=ffffff"></iframe>
However, neither machine learning nor artificial intelligence are isolated when considering global technology developments. Rather, they both belong to one big story which connects "war machines, computer networks, social media, ubiquitous surveillance and virtual reality" - according to Kevin Kelly, the founder of Wired.
(the Amazon advertisement contains an affiliate link for MachineCurve).
If you're interested in either the business or technology aspects of AI and/or machine learning, I would like to recommend the 2016 book Rise of the Machines: the lost history of cybernetics by Thomas Rid, a German professor specializing in political science at Johns Hopkins University in the USA (Wikipedia, 2016).
Rid, who specializes in how politics shape information technology and how IT shapes politics (and society) in return, with Rise of the Machines provides an excellent and thorough overview of the global trends in WW2 and post-WW2 technology developments.
Setting the stage
The book begins in the autumn of 1940, when German fighter pilots raid the city of London, the Londoners being unknowing about how the concept of war would change in the years to come (Rid, 2016). The Battle of Britain would be fertile ground for many technology developments such as radar technology being capable to track enemy aircraft, variable-time fuse shells which explode when near the enemy and so on.
However, it wouldn't stop there. Rather, with scientists like Turing inventing the Bombe - an electromagnetic computer cracking German Enigma codes - a technology revolution was about to begin (Wikipedia, 2001). Rid's story presents how this technology story has unfolded over many decades.
Starting with the movement of cybernetics, which in the later 1940s emerged to make sense of how technology had evolved, Rid shows how through controlling the environment by means of feedback humans and machines had now been coupled tightly at an unprecedented scale.
This is followed by chapters on automation, discussing the long-term effects of such human-machine symbiosis on e.g. employability, on organisms, discussing whether technology could physically integrate with organisms, to culture and space, introducing how technology spawned entirely new subcultures such as cyberpunk as well as new territory called cyberspace.
We think virtual reality is a new development - it's not. It's intrinsically linked to the concept cyberspace discussed in Rid's book. Photographer: Bruce Mars, Pexels License.
But who controls cyberspace? Rid's book continues until today, discussing the battle between anarchists and governments about who owns the vast digital lands, and how the introduction of cryptography has substantially polarized this debate.
Entering today's world, the book discusses war again - digital war indeed, introducing cybercrime and cyberattacks, the type of warfare that is ubiquitous today. In its conclusion, Rid is spot on: "cybernetics started at war - and eventually came back to war" (Rid, 2016). Today's world has digitized and Rid's book tells us how it did.
AI's broader context
Now, why would this book be a recommendation if you're interested in AI?
I get the question - let me explain.
Nothing in this world happens in isolation. Any action is triggered by some previous action and will trigger another action - or perhaps a few of them - which in return spawn more actions, and so on. Hence, I think that it's important to study context when discussing some phenomenom, and preferably as objectively as possible.
The same is true for Artificial Intelligence. Did you know that in the World War 2 era, Turing already undertook thought experiments about AI, questioning whether it was possible - with the Turing test as a prime example? That the ideas about today's narrow AI systems, which is that they often work best when they support humans (i.e., human-machine symbiosis), are grounded in decades-old concepts?
That the ideas put forward by the so-called singularity movement, claiming that superintelligent AI will create an exponentially better world for humans to live in, have been here since the late 1950s?
(And that the same is true for the apocalyptic thoughts about the same superintellitent technology?)
Well, you get the point.
If you wish to understand today's AI developments, you'll have to consider them in the broad context of technological history. Thomas Rid's Rise of the Machines is, although written by an academic and hence sometimes a little challenging to plough through, an excellent chronology of how technology has shaped the world. Absolute recommendation!
Rise of the Machines: the lost history of cybernetics
Thomas Rid, 2016
Wikipedia. (2016, October 20). Thomas Rid. Retrieved from https://en.wikipedia.org/wiki/Thomas_Rid
Rid, T. (2016). Rise of the Machines: the lost history of cybernetics. Scribe Publications.
Wikipedia. (2001, November 12). Alan Turing. Retrieved from https://en.wikipedia.org/wiki/Alan_Turing#Bombe