AI may also exacerbate income and wealth inequality

  Digital technologies are transforming the economy, especially when it comes to artificial intelligence, and these technologies are rapidly changing the way we live and work. But the shift presents a troubling conundrum: Even as income inequality rises, these technologies have done little for economic growth.
  Economists believe productivity growth is critical to raising living standards, but productivity growth has been largely subdued in many countries since at least 2000.
  Why have these technologies failed to bring more economic growth? Why haven’t they driven broader prosperity?
  AI researchers and businesses focus on building Machines to replicate human intelligence.
  The title, of course, is a reference to Alan Turing and his famous 1950 test of whether a machine is intelligent: Can it imitate a person so well that you can’t tell it’s not a person? “Many researchers have pursued this goal since then,” Brinjovson said.

  But, he said, the obsession with mimicking human intelligence has led to AI and automation that often simply replace workers rather than expand human capabilities to enable people to perform new tasks.
  For economist Brin Joffson, while simple automation creates value, it can also lead to rising income and wealth inequality.
  The undue focus on human-like AI has driven down wages for the majority, he wrote, “even amplifying the market power of the few who own and control these technologies.”
  In the article, he argues that the emphasis on automation, rather than its enhancement, is the “biggest single explanation” for the rise of billionaires at a time when the average real wage of many Americans has fallen.
  Brinjovson is not a Luddite (those who oppose improved methods of work, who oppose the use of new machines). He co-authored a book with Andrew McPhee in 2014 called The Second Machine Age: Work, Progress, and Prosperity in an Age of Brilliant Technology. But he says AI researchers are too limited in their thinking.
  ”I’ve talked to a lot of researchers and they said: ‘Our job is to make a machine like a human’ is a clear vision.” But, he added, “it’s also a low standard of laziness. ”
  He believes that in the long run, more value will be created by using AI to produce new goods and services rather than simply trying to replace workers.
  But he said it was often easier for companies, driven by cost-cutting, to replace machines than to rethink processes and invest in technology that leverages artificial intelligence to expand a company’s offerings and increase employee productivity.
  Recent advances in artificial intelligence have been impressive, ranging from driverless cars to human-like language models. However, steering the technology’s trajectory is critical.
  Thanks to the choices researchers and businesses have made so far, new digital technologies have created enormous wealth for those who own and invent them, while also destroying jobs that are easily displaced.
  These inventions created good tech jobs in a few cities like San Francisco and Seattle, while most others were left behind. But it doesn’t have to be this way.
  MIT economist Daron Azimolu provides compelling evidence for the role of automation, robots and algorithms in slowing U.S. wage growth and increasing inequality.
  In fact, between 1980 and 2016, automation accounted for 50 to 70 percent of the increase in U.S. wage inequality between 1980 and 2016, he said.
  This was mainly before the surge in the use of artificial intelligence technology. And Azimolu worries that AI-based automation will make things worse.
  In the early 20th century and earlier periods, shifts in technology often created more good new jobs than they destroyed, but that no longer seems to be the case.
  One reason is that companies often choose to deploy what he and his collaborator Pascual Restrepo call “general technology” that replaces workers but is critical to improving productivity or creating new business opportunities little contribution.
  At the same time, companies and researchers have largely ignored the potential of AI technologies to expand the capabilities of workers while providing better services.
  Digital technology could allow nurses to more accurately diagnose illnesses, or help teachers deliver more personalized lessons to students, Azimolu noted.
  Both AI scientists and big tech companies are gravely responsible for making decisions that favor over-automation, Azimolu said.
  Federal tax policy favors intelligent development. While labor is heavily taxed, there is no payroll tax on robots or automation.
  And, he said, AI researchers “have no guilt about working on technologies that automate jobs at the cost of many people losing their jobs.”
  But he retains the strongest anger at Big Tech, citing figures that show U.S. and Chinese tech giants fund about two-thirds of AI work.
  ”I don’t think it’s an accident that we put so much emphasis on automation, because the future of technology in this country is in the hands of a handful of companies like Google, Amazon, Facebook, Microsoft, and those companies that just use algorithmic automation as their business model,” he said.
rebound effect

  Anger over AI’s role in increasing inequality could jeopardize the future of the technology. In her new book, Gears and Monsters: What Economics is and Should Be, Cambridge University economist Diane Coyle argues that the digital economy requires new ways of thinking about progress.
  “Whatever we mean by economic growth, better things, the gains have to be shared more evenly than they have been recently,” she wrote. Earning jobs that are undercut by automation will be politically unsustainable.”
  To improve living standards and increase prosperity for more people, Coyle said greater use of digital technology is needed to increase productivity across industries, including healthcare and construction.
  But people can’t be expected to embrace the changes if they don’t see the benefits—if they just see good jobs destroyed.

  In a recent interview with MIT Technology Review, Coyle said she worries that inequality in tech could be a barrier to deploying AI. “We’re talking about disruptive issues,” she said. “These are transformative technologies that can change the way we spend our time every day and transform already successful business models.” To make such a “massive change” you need support from society, she added.
  Instead, Coyle said, dissatisfaction is brewing among many because they believe the benefits go to a few elites in prosperous cities.
  In the United States, for example, for most of the 20th century, the various regions of the United States—in the words of economists—were “converging” and financial disparities narrowed.
  Then, in the 1980s, with the onslaught of digital technology, that trend was reversed. Automation has destroyed many manufacturing and retail jobs, and new high-paying tech jobs are concentrated in a few cities.
  Eight U.S. cities, including San Francisco, San Jose, Boston and Seattle, had about 38 percent of tech jobs on the list in 2019, according to the Brookings Institution.
  New AI technologies are particularly concentrated: Mark Murrow and Sifan Liu of the Brookings Institution estimate that just 15 cities account for two-thirds of U.S. AI assets and capabilities.
  The dominance of a handful of cities in the invention and commercialization of artificial intelligence means that geographic disparities in wealth will continue to soar.
  Not only will this spark political and social unrest, but, as Coyle argues, it could also hamper the various AI technologies needed for regional economic growth.
  Part of the solution may lie in loosening the grip of big tech companies on defining the AI ​​agenda in some way. This may require increased federal research funding independent of the tech giants.
  For example, Murro and others have suggested huge federal funding to help create regional innovation centers in the United States.
  A more immediate approach is to broaden our digital imaginations, envisioning that AI technology is not simply replacing jobs, but increasing opportunities in areas of greatest concern to different parts of the country, such as healthcare, education, and manufacturing.
change your mind

  Artificial intelligence and robotics researchers like to replicate human capabilities, which often means trying to get a machine to perform a task that is easy for humans but technically difficult. For example, make a bed or make an espresso. Or a self-driving car driving the streets of a city, or a robot acting as a barista. But all too often, those developing and deploying these technologies fail to consider their potential impact on employment and labor markets.
  Anton Korinek, an economist at the University of Virginia and a Rubinstein Fellow at Brookings University, said that once self-driving cars are deployed, tens of billions of dollars in R&D investment will inevitably have a negative impact on the labor market. Take the jobs of countless drivers.
  What if those billions of dollars were invested in artificial intelligence tools that are more likely to expand workforce opportunities, he asked.
  When applying for funding from agencies like the National Science Foundation and the National Institutes of Health, “nobody asks, ‘How is it going to affect the labor market?'” Kolinek explains,
  policy expert Cards at AI Collaboration in San Francisco Tia Klinova is researching how to get AI scientists to rethink the way they measure success.
  ”When you look at AI research, you see benchmarks that are commonly used that match or compare human performance,” she said. That is, when AI scientists rate their programs, they Image recognition, for example, is by comparing it with a person’s ability to recognize objects.

  These benchmarks drive the direction of research, Klinova said. “It’s not surprising that there’s automation and stronger automation that’s emerging now,” she added. “Benchmarks are very important to AI developers—especially for the large number of people who are entering AI and are confused about ‘Should I What to do?’ for young scientists.”
  Klinova said there is currently a lack of benchmarks for human-robot collaboration performance, although she has begun efforts to help create some.
  The team she collaborates with her AI is also writing a user guide for AI developers without a background in economics to help them understand how workers might be affected by the research they do.
  ”It’s changing the narrative that AI innovators are given a blank ticket for disruption, and then it’s up to society and government to handle,” Krinova said. Every AI company has something to say about it, she said. Answers to AI bias and ethics, “but they still don’t account for the impact on the workforce.”
  The pandemic has accelerated digital transformation. Businesses have understandably turned to automation to replace workers. But the pandemic has also demonstrated the potential of digital technology to broaden our capabilities.
  They have given us the tools to develop new vaccines and a viable work-from-home model for many.
  AI will inevitably expand its impact, and it will be interesting to see if this will result in greater damage to good jobs and more inequality. “I’m optimistic that we can steer this technology in the right way,” said Brinjovson, but he added that it will mean making careful choices about the technologies we create and invest in.

Comments Off on AI may also exacerbate income and wealth inequality
error: Content is protected !!