- Companies that invest heavily in AI are growing faster and hiring more people
- Cities where AI is an economic driver report higher growth and citizen well-being
- The benefits of AI are concentrated in the top 10% of enterprises
At Optum, advanced analytics are just what the doctors are ordering.
The company, a division of UnitedHealth Group (UHG), the second largest U.S.-based healthcare company, says it’s using artificial intelligence (AI) to slash healthcare delivery costs while improving patient outcomes.
Optum’s Case Advisor platform uses natural language processing and machine learning to analyze complex medical records, allowing physicians to spend less time poring over charts and more time treating patients. Its pharmacy services business uses analytics to detect fraud and prevent abuse of pharmacy benefits. An Optum analytics platform helps caregivers understand patient populations better and intervene earlier, altering behavior that could lead to chronic conditions or costly visits to the emergency room.
“AI is incredibly important to the future of healthcare,” says Sanji Fernando, Optum’s senior vice president for AI and analytics platforms. “By leveraging the benefits of machine learning and AI, we can improve efficiency, lower costs, and create new products and services.”
It’s also been very good for Optum’s bottom line. The division recorded revenues of $136 billion in 2020—more than half of UHG’s annual total—for a year-over-year growth rate of 21%.
Optum’s success is one of the examples cited in a 2020 study, “Artificial Intelligence, Firm Growth, and Industry Concentration,” that arrives at a striking conclusion: Instead of eliminating employees through automation, companies that invest the most in AI products and services are growing faster, expanding further into new markets, and hiring more people than companies that haven’t.
Other recent studies also point to the positive effects of widespread AI adoption. That’s somewhat of a surprise, given that economists have been divided for years about how AI would impact hiring, productivity, and other economic benchmarks.
For the most advanced firms, researchers found, investments in AI result not in fewer jobs but in job growth. Major U.S. metro areas where AI is a significant driver of the local economy are not only growing faster, they are seeing a rise in overall well-being, according to another recent study by researchers at the Massachusetts Institute of Technology and Stanford University.
And as Optum shows, AI is transforming health care in important ways; AI investments in drug discovery have increased fourfold over the past year, according to the 2021 AI Index published in March 2021 by Stanford’s Institute for Human-Centered Artificial Intelligence (HAI).
“AI has huge potential to help people, especially with healthcare,” says Erik Brynjolfsson, director of Stanford’s Digital Economy Lab and a senior fellow at HAI. “This is a great example of how technology can help us live longer, healthier lives.”
[Read our Q&A with Brynjolfsson: Automation and the ‘winner take most’ effect]
Automation fosters growth
Last September, researchers from UC Berkeley, Columbia University, the University of Maryland, and the AI for Good Foundation published the results of a study gauging how AI investments impacted job growth and industry concentration. Using AI-related job postings and online resumes as a proxy for investment, the team measured how the technology affected productivity, sales, and employment at publicly traded companies.
The study’s conclusion: “Firms that invest in AI grow more.”
Firms are expanding the boundaries of their geographic locations and their product offerings by innovating with AI technology.
Companies with the greatest increase in the number of AI workers saw an 11% to 15% increase in overall job growth and sales, and roughly a 1% boost in market share, between 2010 and 2018.
The greatest benefit from AI was the ability to expand into new markets and introduce new products, notes study co-author Anastassia Fedyk, an assistant professor of finance at Berkeley’s Haas School of Business.
“Firms are expanding the boundaries of their geographic locations and their product offerings by innovating with AI technology,” she says.
To avoid biasing the results toward companies with unusually high investments in AI, the study deliberately excluded tech giants such as Google and Facebook, says co-author James Hodson, director of the AI for Good Foundation. “We’re really measuring the impact on mainstream business,” he says.
Still, enterprises with the resources to invest in AI are going to reap the greatest rewards from the technology, Hodson adds. These companies, which tend to have a more established data infrastructure, are more likely to grow faster, achieve greater economies of scale, and gain efficiency advantages. If that’s the case, over the next several years AI may give early adopters a competitive edge over those that come later to the game, although Hodson notes that isn’t visible yet in their research.
“The largest, most productive firms are also the ones that tend to have more resources,” he says. “If AI takes a big upfront investment to do it at scale, these are exactly the places where you would expect to see this growth happen.”
More AI, better living?
An August 2020 study by researchers at Stanford and Arizona State University found direct links between AI investment and economic growth. Cities where companies are hiring more AI workers enjoy more robust economies, notes study co-author Saurabh Mishra, manager of the AI Index Program at Stanford’s HAI.
“AI jobs are really a proxy for the digital revolution, broadly speaking,” says Mishra. “If you have a high concentration of AI jobs, there’s a good chance you’re living in an area that’s more tech-oriented.”
People living in those cities also registered higher levels of well-being, as measured from daily phone surveys conducted by Gallup, the study found. AI’s relationship to better living, however, may be more a function of correlation than causation, as people working in and around AI-powered jobs likely also have higher average salaries and perks that contribute to greater career satisfaction and better health.
But the results also suggest that doubling the share of AI jobs in a city would result in a nearly 5% boost in overall well-being, says co-author Christos Makridis, an assistant research professor at Arizona State University and a fellow at Stanford’s Digital Economy Lab.
AI’s ability to automate repetitive tasks, says Makridis, makes a person’s job more interesting.
“By reducing busywork and distractions, managers can focus employees more on value-added work,” he says, “which is associated with greater job satisfaction and meaning.”
“Winner take most”
The benefits of AI, though, aren’t evenly distributed, and digital technologies are creating so-called “superstar firms,” according to studies conducted by researchers at the Wharton School of Business and the Stanford Digital Economy Lab.
The greater the amount of digital capital an enterprise has accumulated, the larger the productivity gains it will see within three years.
In a paper published in December, researchers calculated the impact of “digital capital”—products and services spurred by mobile technology, cloud computing, big data, analytics, and AI, as well as the process and organizational transformations required to take advantage of them.
The greater the amount of digital capital an enterprise has accumulated, the larger the productivity gains it will see within three years, the study concludes. This finding holds true across all industries, not just in technology, says Brynjolfsson, who is one of the study’s co-authors.
The key differentiator between the digital haves and have-nots is their leadership’s willingness to embrace new technologies.
The divide between the haves and have-nots extends to AI startups. The amount of private investment in AI startups increased almost 10% last year, according to the Stanford AI Index. But the number of startups receiving that money decreased for the third consecutive year, contributing to an increasing gap between winners and losers.
“We’re seeing this ‘winner take most’ effect,” says Brynjolfsson, “where a few firms are dominating the adoption of these technologies and getting the biggest market value gains from them.”
Most of the benefits of this digital transformation have been concentrated in the top 10% of enterprises. That’s a warning signal to the others. “The sooner those companies act,” Brynjolfsson says, “the less likely they’ll get left behind.”