Top AI experts think job displacement is inevitable but beneficial, and propose ideas for addressing economic inequality
Andrew Ng, Kai-Fu Lee, and Sebastian Thrun discuss the future of work
A powerhouse panel of artificial intelligence experts was assembled by Sinovation Ventures last week to discuss the future of AI. The panelists included Sinvoation Ventures and Google China founding president Kai-Fu Lee, Google Brain founder Andrew Ng, and Udacity and Google X co-founder Sebastian Thrun.
Job displacement is a question of when, not if
The pace of AI development is increasing. Lee attributes the growth of AI to the multiplicative effects of having “the number of fields that have enough data for existing algorithms to work” in addition to “new algorithms that will be invented.” Ng indicated that supervised learning, in particular, has clear potential to transform a wide range of industries.
With this rise of AI has come increased apprehension about the societal and economic effects of these new technologies. Phrases like IA — intelligence augmentation/amplification — and “human-in-the-loop” machine learning have become popular ways to describe artificial intelligence technologies in a less threatening manner. And these ideas are valid; Thrun argues that the “entire history of technology is IA” and that advances from farm tractors to telephones have complimented people in ways that that have made us “superhuman.”
Nevertheless, one thing was clear to all the panelists — considerable job displacement is on the horizon. The first to go? Office workers who perform repetitive tasks.
“75% of us in America work in offices,” According to Thrun. “Almost all of us do extremely repetitive work … even the CEO. When you have AI looking over your shoulder, I believe anything repetitive in offices can be automated.”
For these office workers, he seemed to see no place for IA. “Most of the replacement — traders on Wall Street, call center agents — they’re completely going to be gone,” Thrun added.
Similarly, Ng, who has spent time interviewing and understanding the work of call center agents stated bluntly that his “teams can write software to automate a large fraction of these jobs away.”
In his keynote speech, Lee even outlined a timeline of job displacement that he believes we will see over next 15 years. He predicts that AI technology will replace repetitive work (tellers, receptionists, customer service) in 3–5 years; jobs like traders, analysts, and assistants in 5–10 years; and drivers, radiologists, and engineers in 10–15 years.
Elimination of boring jobs will clear the way for greater human creativity
Despite this seemingly harsh outlook, the panelists all shared an excitement for the benefits they saw from automating repetitive work. In particular, they it as a potential remedy for the slowdown of labor productivity that has been observed over the last several years.
Lee speculated that “human productivity is flat because the jobs are boring” and that more exciting jobs would naturally motivate people to be more passionate and productive in their work.
Thrun echoed this sentiment towards “boring” jobs. “I can’t wait for these stupid jobs to go away,” he said. “I think some of these jobs are very degrading.”
In addition to enabling an increase in productivity, Thrun is confident that AI can bring about flourishing of human creativity. Drawing a parallel to the machines of the Industrial Revolution, which he claims allowed for greater free time and led to increased literacy rates, Thrun claims that AI can free people to do more meaningful work. “This is the beginning of human creativity. I think there’s so much creativity to be unleashed. I just can’t wait until we stop wasting our time.”
“AI is going to be to the office worker the same way the Agricultural Revolution and the steam engine has been to farmers.” — Sebastian Thrun
Given that AI has surpassed human capabilities in many areas, Lee repeatedly emphasized what he saw as the long-term human advantage — the human heart. He maintained that jobs that involve human-to-human interactions (bartenders, masseuse, concierge) and creativity (artists, writers) will still benefit from the “human touch” and “a little love in the process;” these jobs are likely to thrive alongside AI.
“AI is getting better than people, but maybe we need to go back to our roots and think more about our heart and what we can do as humans for other human beings, making the world a better place.” — Kai-Fu Lee
Wealth redistribution policies should be considered to combat inequality
In order to bridge the gap between job displacement and productivity growth, Ng emphasized the need for solutions involving education “in order to re-skill very large segments of our population.”
Without such solutions, AI automation technologies may exacerbate the problem of wealth inequality. In addition to rethinking education, the panelists also proposed other fairly concrete ideas for wealth distribution.
Lee suggested an approach he describes as “taxation and minimum stipend,” in which large tech companies and wealthy corporations are taxed and the revenue is used to impose a higher minimum wage or living stipend for those who may have been displaced by automation technology.
In addition to a negative income tax and universal basic income, Ng called for a “new New Deal.” He proposed a “conditional basic income, where we pay people but where the payment to unemployed individuals is tied to their studying. Because by studying, you increase your chance of re-entering the workforce and contribute back to the taxpayers.” He saw this as a better structure of incentives for education.
Wealth redistribution may also be necessary on a global level. Lee made sure to include developing nations in the conversation by pointing out that “poorer countries won’t have the rich companies to tax to redistribute the income to get re-education going. So, some kind of redistribution at the country level may be needed. Put it very directly, they basically get the U.S. or China to subsidize them, or maybe there are other ways through the United Nations.”
On the bright side, technology is also acting as an equalizing force
In the long run, the panelists viewed technology as a way to democratize access to many key resources. Education, food, transportation, shelter, and health are all being made more affordable through technology, according to Thrun.
“And these things, I think they are not dispersed equally, but they are dispersed worldwide at this point,” Thrun said. “I think there’s a lot of leveling going on in the world that I see as very positive.”
In particular, both Lee and Ng pointed to greater access to education, enabled by “the rise of digital content” and “smart AI algorithms [for] teaching,” as a key equalizing force.
As evidenced by the user demographics of Coursera, Ng claimed that we are now “better than ever at giving those opportunities to people around the world, including developing economies.”
