Erik Brynjolfsson – The Second Machine Age

ERIK BRYNJOLFSSON:  Hi.  I’m Erik Brynjolfsson and I’m here to talk to you about the Second Machine Age.  To change the world, you need to do two things.  You need to have the power to physically move atoms around, but you also need to have the intelligence to know where to put them.  You need a power system and a control system.

The first Industrial Revolution was mostly about creating much better, more effective power systems like the steam engine.  This time around, it’s focused more on cognitive tasks/mental tasks.  It’s not as clear whether humans will be complementary or perhaps substituted by these new technologies.  Technology has always been destroying jobs.  It’s always been creating jobs, but that pace is speeding up.

Consider the example of a simple tax preparation program like Turbo Tax.  This software is used by tens of millions of people.  The Founders of Intuit are now billionaires with hundreds of thousands of people who were charged with doing taxes.  They’re not very valuable anymore because the software can do it much more cheaply and often more accurately, and more easily.  This is a microcosm, which is happening in more and more industries.  It’s entirely possible for technology to be racing ahead, creating an enormous wealth but at the same time, there’s no economic law that says that everybody’s going to share in that wealth or even the majority of people.  The right strategy is to speed up our organisations, our skills, and our institutions as they adapt to those new technologies.

When the Industrial Revolution automated a lot of agricultural work, all those farmers didn’t simply become unemployed.  They found new industries but that required new skills.  I’m hopeful that digitisation can help with the reinvention of education.  The great promise of these technologies is not just that they can take the best course materials and replicate them to more and more people, but secondly and perhaps more importantly, they provide an enormous amount of data.  We can learn which educational techniques are working and which ones aren’t, and customise and provide individual feedback to each of them.

That will not only raise the level of education, but also increase the slope of education so that we’re learning how to learn faster and faster.

You’d be surprised how often I run into economists who worry that we are running out of innovations.  We describe the concept of combinatorial innovation, the idea that most innovations are created when you combine two previous innovations.  That means that instead of using up innovations, each innovation creates building blocks for more and more innovations.

An undergraduate student of mine wrote a simple app on Facebook in a matter of two or three weeks.  Within a few months, he had over one-point-three million users.  The reason he was able to do it so easily was because he built it on top of Facebook and Facebook was built on top of Tim Berners-Lee’s Worldwide Web, and that was built on top of the Internet so I’m not worried about running out of innovations.  I think that every time we invent something, we make it easier to invent other things, using the previous technologies.

When agriculture went from 90 percent of the population to two percent of the population today, inventors and entrepreneurs like Henry Ford, Steve Jobs, Bill Gates, and many others helped create several entirely new industries that didn’t exist before.  We need to do the same thing today.  We need to figure out what the next new jobs are going to be.

Right now, we have a set of taxes on low wage labour that effectively discourage employers and entrepreneurs from creating new businesses that hire a lot of people.  Instead, we would eliminate some of those taxes and even subsidise the employment of people in those areas.  We’d be more likely to have entrepreneurs investing and inventing industries that employ a lot of people.

There are three classes of recommendation that we make.  The first has to do with education, the second with entrepreneurship, and the third has to do with tax fiscal policy.  We can’t simply double down on the old strategy of more mass education.  We need to reinvent education because the kinds of skills that are being taught in the Industrial Age aren’t necessarily the ones we need going forward.

We need more creativity.  We need more interpersonal skills, and that requires a new kind of education.  It’s also important to encourage more entrepreneurship.  Now you might think we already have a lot of entrepreneurship in places like the United States and it’s true.  We do have a fair amount but we’re not doing it fast enough because we’re not creating jobs as rapidly as we’re automating the old ones.

There’s also a lot we can do with tax policy.  The earned income tax in the United States is a small example of that.  It helps raise wages of low-income people without penalising the employers.  Instead, it encourages the employers to hire more of them.  If we combine these policies – the educational, the entrepreneurial, and the tax policy – we’ll be able to overcome these challenges, just as we overcame some of the challenges of the First Machine Age.

My final recommendation to you – and it’s really more of a challenge – is to learn from what was so successful in creating the Google car and other technological breakthroughs.  The Google car was a response to what was called the DARPA Grand Challenge.  We need a new Grand Challenge that’s focused not on motivating the technologists and inspiring them, but on transforming our economic system and our social system so that we can reinvent that and have it match the new technologies of the Second Machine Age.

Visited 218 times, 1 visit(s) today
Categories WEF