Stanford and MIT study: A.I. boosted worker productivity by 14%—those who use it 'will replace those who don't'

Story by Jennifer Liu • 12h ago

Artificial intelligence tools like chatbots helped boost worker productivity at one tech company by 14%, according to new research from Stanford and MIT that was first reported by Bloomberg.



The study is thought to be the first major real-world application of generative AI in the workplace. Researchers measured productivity of more than 5,000 customer support agents, based primarily in the Philippines, at a Fortune 500 enterprise software firm over the course of a year.

Tech support agents who used AI tools that created conversational scripts boosted their productivity, measured as issues resolved per hour, by 14% on average, but the improvement was even more pronounced for "novice and low-skilled workers" who were able to get their work done 35% faster.

In some cases, using AI trumped having real-life work experience: Customer service agents with two months of experience who used AI support performed as well or better than agents with over six months of experience working without AI.

Meanwhile, the use of AI tools showed a minimal impact on "experienced or highly skilled workers," the authors note, and at times served as a distraction.

AI support can be especially helpful to entry-level or early-career workers, says Lindsey Raymond, an MIT Ph.D. candidate and co-author of the paper. Less experienced workers benefit from AI by taking its recommendations to get up to speed and learn skillsets that usually come with experience.

With that said, AI tools benefit from the best and brightest workers training the AI itself by providing examples of best practices, which the technology then turns into recommendations for others workers to apply.

Businesses should understand that, despite less dramatic changes in productivity, high-performing employees should be recognized and compensated for generating the solutions that others can learn from, Raymond says.
Happier workers and customers

The year-long experiment also revealed that AI assistance improved customer satisfaction, reduced requests for managerial intervention and improved employee retention.

The research isn't meant to hypothesize whether AI will replace workers, Raymond says, but rather concludes the technology will help workers more effectively multitask and handle more complicated questions faster.

Better and faster work led to happier customers, who were in turn nicer to customer service agents and improved employee retention, Raymond says.

Tools that make people more effective at their jobs make the experience of work less stressful, she adds.

Results that generative AI can boost productivity is generally good news, though the biggest benefits may not be evenly distributed, says Erik Brynjolfsson, the director of the Digital Economy Lab at the Stanford Institute for Human-Centered AI, and co-author on the report. "There's no guarantee we'll all benefit, but it certainly sets the table for us all being better off," he says.

'Workers who work with generative AI will replace those who don't'

Brynjolfsson says call centers are a great place to use generative AI because it involves a lot of scripted language, but that "almost any kind of information or knowledge work that involves language could benefit from this," including across legal, marketing, medicine, teaching and other fields.

Workers at all levels can benefit from the technology, he adds — he recently spoke with a CEO who used generative AI to prepare for a board meeting.

"Probably over half of the U.S. workforce will be significantly affected by these tools," Brynjolfsson says.

He adds that workers, especially young workers, can stay ahead of the curve by embracing the reality of the technology: "Workers who embrace the technology, play around with it and learn how to use it are the ones that are going to succeed and benefit the most," Brynjolfsson says. "I don't think the generative AI is going to replace workers, but workers who work with generative AI will replace those who don't."

Some experts say generative AI tools could affect how two-thirds of current jobs are performed and could eventually raise global gross domestic product by as much as 7%, according to one economic report from Goldman Sachs.

One recent survey of LinkedIn's Top Companies found that nearly 70% say AI is already helping them be faster and smarter, and another 32% say they expect to see larger gains from using AI in the coming years. And companies like EY explicitly listed AI as one of their top three hiring priorities, while Wells Fargo and Kaiser Permanente are implementing AI across their workflows.

Check out: Meta has a new AI tool to fight misinformation—and it’s using Wikipedia to train itself