2012 / Trend # 1 / Algorithms are the new machines

I'm currently reading a book by Douglas Hofstadter called "Godel, Escher, Bach". The title and the cover of the book are enough I guess to give you an idea of how mind-boggling this book is

At page 60, Hofstadter lays down the question that will occupy him for much of the rest of the book:

In Chapters to come, we will lay out a formal system that ( 1 ) includes a stylized vocabulary in which all statements about natural numbers can be expressed, and (2) has rules corresponding to all the types of reasoning which seem necessary. A very important question will be whether the rules for symbol manipulation which we have then formulated are really of equal power (as far as number theory is concerned) to our usual mental reasoning abilities-or, more generally, whether it is theoretically possible to attain the level of our thinking abilities, by using some formal system.

This last sentence "whether it is theoretically possible to attain the level of our thinking abilities, by using a formal system" is probably the the most important question of the next century. To what extent can we replicate the human mind's functioning and abilities ? And hidden in this very question however is the preliminary step that is driving much of today's innovations and design thinking : Can we formalize our mental processes ? And since mathematics is the language we are using to accomplish this formalization and feed it to computers, the question translates into : Is it possible to express every single human mental ability in terms of a mathematical algorithm ?

Now beforehand, I had read Charles Munger's 1994 speech to the USC Business School graduates “A Lesson on Elementary, Worldly Wisdom As It Relates To Investment Management & Business”. This isn't a random speaker by the way : Charlie Munger is Warren Buffet's "secret weapon". He is probably one of Berkshire Hathaway's most valuable assets and one of the greatest investment minds today. In his USC speech, he actually explains how he thinks ... You read that right : It's worth a read. This single quote summarizes the whole thing though :

What is elementary, worldly wisdom?  Well, the first rule is that you can't really know anything if you just remember isolated facts and try and bang 'em back.  If the facts don't hang together on a latticework of theory, you don't have them in a usable form.

You've got to have models in your head.  And you've got to array your experience ‑ both vicarious and direct ‑ on this latticework of models.  You may have noticed students who just try to remember and pound back what is remembered. Well, they fail in school and in life.  You've got to hang experience on a latticework of models in your head.

What are the models?  Well, the first rule is that you've got to have multiple models ‑ because if you just have one or two that you're using, the nature of human psychology is such that you'll torture reality so that it fits your models, or at least you'll think it does. You become the equivalent of a chiropractor who, of course, is the great boob in medicine.

To be clear however, mental models, which in a way are Munger's equivalent of Hofstadter's formal systems, are not the way we think. They are probably, however, the best way we have to think, analyze data and reach decisions. Indeed, our brains are responsive to stories and in its essence, the latticework of models Munger suggests we use is a story-building stratagem.

Models or formal systems are interesting however not only because they are efficient but also because they demonstrate a certain ability to replicate the world. Nowadays, algorithms are also one such attempt to do so in the realm of human thinking. Here's some insight into an upcoming trend:

  1. Economist William Brian Arthur recently published an article in McKinsey Quarterly about "The second economy". An economy parallel to ours where algorithms have taken everything in charge. Whereas humans used to handle your airplane ticket reservation, dozens of algorithms talk to each other in order to make that reservation nowadays. Arthur argues that this might raise unemployment in the long term. That's in part where this blog's title gets its inspiration from: Just as machines replaced workers in the industrial sector, algorithms are about to replace workers in the services sector.
  2. A Wired article titled "Algorithms take control of Wall Street" goes further and explains that not only are we being kicked out of the system by algorithms but moreover that this trend is here to stay : "Today Wall Street is ruled by thousands of little algorithms, and they've created a new market—volatile, unpredictable, and impossible for humans to comprehend" and "We may be able to slow it down, but we can never contain, control, or comprehend it. It’s the machines’ market now; we just trade in it."

Arthur's article is interesting in that he understands that in a world where algorithms are taking over our jobs, we need to imagine a new world where prosperity and revenue distribution would be reconfigured. Wired's article reminds us of the imperfections of our inventions. Just like us, algorithms are not infallible :

For individual investors, trading with algorithms has been a boon: Today, they can buy and sell stocks much faster, cheaper, and easier than ever before. But from a systemic perspective, the stock market risks spinning out of control. Even if each individual algorithm makes perfect sense, collectively they obey an emergent logic—artificial intelligence, but not artificial human intelligence. It is, simply, alien, operating at the natural scale of silicon, not neurons and synapses.

That is how we ended up with the May 6 flash crash :

The May 6 flash crash. The culprit, the report determined, was a “large fundamental trader” that had used an algorithm to hedge its stock market position. The trade was executed in just 20 minutes—an extremely aggressive time frame, which triggered a market plunge as other algorithms reacted, first to the sale and then to one another’s behavior. The chaos produced seemingly nonsensical trades—shares of Accenture were sold for a penny, for instance, while shares of Apple were purchased for $100,000 each. (Both trades were subsequently canceled.) The activity briefly paralyzed the entire financial system.

And an excellent TED talk by Kevin Slavin about the subject (I know I'm always linking to these :) but they do tend to be exhaustive) :

[ted id=1194]

Will this end in a Terminator-style algorithm-controlled robots versus humanity ? I think/hope we're smart enough to avoid unpleasant surprises :) More seriously I believe the only limit to :

  1. Algorithms' development
  2. Their replication of human abilities
  3. Their conquest of the services sector

is the question Hofstadter asks : "whether it is theoretically possible to attain the level of our thinking abilities, by using a formal system". And this greatly echoes with an article I read not long ago : Neuroscience versus Philosophy : Taking aim at free will. An awesome read about how some neuro-scientists are challenging the idea of free will. One interesting experiment shows how our decision to move a hand is determined seconds before we actually take the decision to move our hand. Neuro-scientists spot parts of the brain, distinct from the decision-taking parts, flaring up beforehand.

What if free will does not exist ? The question is most probably an out-stretch but shouldn't be ignored. Especially in our case. If free will doesn't exist, it means our mental frameworks can be entirely computed and there is simply no limit to what algorithms can do and consequently to how obsolete and redundant we might become.

I'm serious here ! A moral and ethics algorithm is perfectly imaginable, a motherly love algorithm implanted into a robot too (don't get too lyrical about these things) ... possibilities are infinite if and only if we can figure out a formal system to replicate our mental processes. Till then let's hope that these replicas/algorithms we'll have created won't sum up to a latticework with a singular ability to out-smart us.