Subscribed ’14: Andy McAfee & The Coming Robot Apocalypse

June 4, 2014




By Gabe Weisert, Content Marketing Manager


Subscribed ’14 continues to hum along, with plenty of packed panels and lively chatter in the hallways. After some festivities last night, the second day of the conference kicked off this morning with a smart, compelling keynote from Professor Andy McAfee that was equal parts fascinating and terrifying.


McAfee is currently a principal research scientist at the Center for Digital Business in the MIT Sloan School of Management. He and Erik Brynjolfsson are co-authors of the book “The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies,” which all Subscribed attendees will be receiving.


He started with a great question: Given that technology is rapidly disrupting huge swathes of the labor force, where do humans still have a sustainable competitive advantage?


The general consensus is that we can still beat the robots fairly consistently in two areas: pattern recognition and complex communication.


The bad news is that we’re losing on both fronts.


Just ten years ago, in their book “The New Division of Labor,” the economists Frank Levy and Richard Murnane cited driving a car as a key example of competitive human visual and sensory pattern recognition.


So much for that.


McAfee had a chance to try out Google’s self-driving car. He described his reaction as having three basic stages: 1. Raw, abject terror prompted by the act of letting go of a steering wheel on the freeway. 2. Passionate interest in the automated driving processes at hand. 3. Utter boredom at the predictable, DMV-approved way the car drives. No one will be driving in five years.


As for complex communication, as a writer I was particularly freaked out to hear about a new algorithm that can take a batch of raw data and turn it into a crisp, elegant prose piece with an active voice, clean narrative arc and supporting arguments. The robots aren’t writing novels yet, but they can handle press releases no problem.


The list goes on: IBM”s supercomputer Watson wiping the floor with poor Ken Jennings on Jeopardy (McAfee described it as watching the “remorseless steamrolling of humanity”), Deep Blue stomping Gary Kasparov at chess, new cancer analytic solutions that could potentially put oncologists out of work, etc.


The bottom line is that technology is eating away at many of our fundamental human competitive advantages, both in the workforce and elsewhere.


So how do we fight back? Or rather, how do we leverage the coming robot army, rather than being run over by it?


With Geeks, Machines and Outsiders.


A Geek can be described as  someone who forms a hypothesis, and then goes out and tests it with the goal of proving themselves wrong. Healthy skepticism is a good thing, particularly in corporate environments.


Geeks can be contrasted with HIPPOS, or the Highest Paid Person’s Opinions. HIPPOS tend to devalue data over concepts like “instinct, experience, and expertise.”


Machines, or analytic engines, give the Geeks the data they need to make informed, quantified judgments.


Outsiders are essentially Geeks that lack a specific vertical expertise.


Large enterprises in particular need to be wary of myopic mindsets that favor domain experience over outside analysis.  No one likes to say “our business is just like all the others.” Everyone likes to believe that their organizations are deeply unique and idiosyncratic.


In short, they like to think that they are experts.


But in  a recent survey of 136 case studies of expert vs. algorithmic predictions, experts batted roughly .059 against the algorithms (!).


Management teams needs to be more open to data-driven analysis that doesn’t necessarily come from years of industry experience. Witness the success of the “geek war room” of the 2012 Obama campaign (presidental campaigns are notoriously HIPPO-friendly environments) and the data-driven predictions of Nate Silver.


And while we need to take more advantage of analytic engines, we also need to establish platforms that place a premium on recurring relationships that provide ongoing customer value. We need to be able to iterate and scale those solutions, not just treat them as one-off transactions (sound familiar?).


Here at Zuora we’re at the forefront of a transition towards recurring revenue relationship business models that many companies interpret as disruptive, even threatening. We’re implementing new business models, and emphasizing new metrics to measure them.


But as J. B. Wood noted in his keynote last night, the top 50 tech companies have lost a combined $200 billion in revenue over the last three years. Software is eating the world. Companies need to reinvent themselves.


By augmenting their existing “system of record” infrastructure with a “system of innovation” Relationship Business Management solution, we’re helping businesses reinvent themsleves in ways that make their value proposition much harder to commodify and automate.


So be nice to Geeks bearing Machines. As our keynote guest speaker Ben Shields of Deloitte said, “Reinvention is a hell of a lot more fun than disruption.”