Happy Nice Mall

In the United States, for example, various parts of the country were “converging”—in the language of economists—for most of the 20th century, and financial inequalities were reduced. Then came the onslaught of digital technologies in the 1980s and the trend reversed. Automation has eliminated many manufacturing and retail jobs. New, well-paid tech jobs are clustered in a few cities.

According to the Brookings Institution, a short list of roughly eight American cities includes San Francisco, San Jose, Boston, and Seattle. 38% of all technical jobs by 2019. New AI technologies are particularly concentrated: Brookings’ Mark Muro and Sifan Liu say it’s not just 15 cities account for two-thirds of AI assets and talent in the United States (San Francisco and San Jose alone account for about a quarter).

The dominance of a few cities in the invention and commercialization of AI means that geographic disparities in wealth will continue to increase. This will not only spur political and social unrest, but may thwart the artificial intelligence technologies needed for regional economies to grow, as Coyle suggests.

Part of the solution may be to somewhat loosen Big Tech’s pressure on defining its AI agenda. This will likely require increased federal funding for independent research from tech giants. Muro and others have proposed heavy federal funding to help create it. US regional innovation centerse.g.

A more immediate response is to broaden our digital imagination to envision artificial intelligence technologies that not only replace jobs, but expand opportunities in sectors that are most important to different parts of the country, such as healthcare, education, and manufacturing.

changing ideas

Artificial intelligence and robotics researchers’ fondness for copying humans’ abilities often means trying to get a machine to do a task that is easy for humans but daunting for technology. For example making a bed or an espresso. Or driving a car. It’s incredible to see an autonomous car cruising a city street or a robot pretending to be a barista. But too often, the people who develop and implement these technologies don’t give much thought to the potential impact on jobs and labor markets.

Anton Korinek, an economist at the University of Virginia and a Rubenstein Fellow at Brookings, said that the tens of billions of dollars spent on the production of autonomous cars will inevitably have a negative impact on labor markets once such vehicles are introduced, and that such vehicles will have countless drivers. What if invested in AI tools that are more likely to expand their opportunities?

“How will it affect the labor markets?” he doesn’t ask,” he explains.

Please consider to support MIT Technology Review journalism subscribe to.

Katya Klinova, policy expert Artificial Intelligence Partnership in San Franciscois working on ways to get AI scientists to rethink the way they measure success. “When you look at AI research and look at pretty much universally used metrics, it all comes down to matching or comparing human performance,” he says. So, AI scientists rate their programs on image recognition, for example, based on how well a person can identify an object.

Klinova says such comparisons are driving the direction of the research. “It’s not surprising that what’s emerging is automation and more powerful automation,” adds she. “The benchmarks are for AI developers, especially those who are collectively entering AI and the ‘What should I be working on?’ It’s very important for young scientists who ask.”

But Klinova says metrics for the performance of human-machine collaborations are lacking, although some have begun work to help create them. Collaborating with She korinek also writes that she and her team at Partnership for AI A user guide for AI developers Those without an economic background to help them understand how workers might be affected by their research.

“It’s about moving the narrative away from one where AI innovators are given a blank ticket to disrupt it and then left to society and government to deal with it,” says Klinova. Every AI firm has some sort of answer about AI bias and ethics, she says, “but they’re still not there for workforce implications.”

The pandemic has accelerated the digital transition. Businesses, understandably, have turned to automation to replace workers. But the pandemic has also pointed to the potential of digital technologies to expand our capabilities. They’ve given us research tools to help create new vaccines and a convenient way for many to work from home.

As AI inevitably expands its influence, it will be worth watching to see if this does even greater harm to good jobs and greater inequality. “I am optimistic that we can steer the technology in the right way,” says Brynjolfsson. But he adds that it will mean making informed choices about the technologies we create and invest in.

Reviewed

“The Turing Trap: The Promise and Danger of Human-Like Artificial Intelligence”
Erik Brynjolfsson
Daedalus, Spring 2022

“The wrong kind of AI? Artificial intelligence and the future of labor demand”
Daron Acemoglu and Pascual Restrepo
Cambridge Journal of Regions, Economy and Society, March 2020

Wheels and Monsters: What Is the Economy and What Should It Be
Diane Coyle
Princeton University Press