Artificial FinTelligence

In his excellent book “Surviving A.I.”, author Calum Chace breaks down the origin story of the Artificial Intelligence ecosystem. The term Artificial Intelligence was first coined in the summer of 1956, by scientist John McCarthy at a month-long conference at Dartmouth College, the foundational event in the genesis of the science.

For the uninitiated, A.I. is a branch of computer science whose stated goal is to create “intelligent machines”, capable of mimicking the human decision-making process across any given range of endeavors. IBM’s Deep Blue computer, which beat Russian chess master Garry Kasparov in 1997, is one of the most famous demonstrations of the early iteration of AI.

Coming into this century, the next phase of A.I. came to the fore: machine learning. In ML, machines “learn” by themselves via the ingesting of large datasets. This allows them to improve with experience. IBM’s Watson beat human players on Jeopardy in 2011 utilizing ML, by being able to crunch 200 million pages of information that had been fed into it during its training. Upon losing, human contestant Ken Jennings famously quipped, “I for one welcome our new robot overlords.” The most potent phase of A.I.’s evolution, illustrated by Google’s AlphaGo beating Go master Lee Sedol, is deep learning. Deep learning is where multilayered neural networks are exposed to vast hordes of data, training machines to solve any problem which requires “thought”.

Cosmic Transformations

A.I. and leveraging machine learning to turn large sets of data into insight is arguably the greatest technological opportunity of all time. For the majority of us, the most obvious demonstration of A.I. sits in our pockets – our smartphones. Outside of smartphones, however, right now A.I. is capable of some staggering feats.

Deep learning neural networks – which are created to mimic the neural connections in our brains and are capable of learning by sucking in huge volumes of data – can execute speech and image recognition at levels higher than us. They can also compose and play music, conduct medical diagnoses, get you off a parking ticket.

The range of A.I.’s application stretches beyond our humble planet, too. In a speech at WIRED2016, a Swiss scientist and “Father of Deep Learning” Jürgen Schmidhuber stated that by 2050 “A.I.’s will colonize and transform the entire cosmos,” with trillions of self-replicating robot factories dotted on the asteroid belt. A “few million years later,” he continued “A.I. will colonize the galaxy”.

PwC ranks A.I. as the biggest commercial opportunity in today’s world, estimating it’ll add $15.7 trillion to the global economy ($6.6 trillion in labor productivity and $9.1 trillion in increased consumer demand), by 2030. To put that into some context, that’s equivalent to:

1. 10x Russia’s GDP
2. 18x Apple’s market cap
3. 186x Bill Gate’s net worth
4. 4132x One World Trade Centers

On the macro front, China aims to lead the world in the field of A.I. by 2030 (and they’re off to a prodigious start). On the micro-level, for the company that leads the AI cloud (democratizing AI as pay-per-use services), it will become the most powerful company in history.