ChatGPT and its AI powers might assist writers and workplace employees enhance their writing high quality and reduce the time spent on duties, slicing out busy work in favor of higher, extra productive work. However, the MIT examine that advised these conclusions additionally warned that employers might use AI to extend layoffs, too.

The paper, “Experimental Evidence on the Productivity Effects of Generative Artificial Intelligence” by Shakked Noy and Whitney Zhang of the economics division at MIT, is taken into account a working paper and has not been peer-reviewed. Still, the conclusions it discovered about ChatGPT’s AI chatbot technology are each fascinating—and troubling, particularly when the examine factored in the way it impacts employees.

The two doctoral college students break up 444 college-educated professionals into two teams, and assigned them to jot down press releases, e mail, brief experiences, and evaluation plans—a traditional workday for many individuals. One of the teams was allowed to make use of a textual content editor. The different was skilled and allowed to make use of ChatGPT to do the work for them. The output was rated, assigned a price rating, and people with the best rating had been paid an precise (although nominal) sum of money for his or her excessive scores, pushing them to ship their finest work.

The examine discovered that those that used ChatGPT reduce the time wanted to finish the duties by nearly half (half-hour to about 17 minutes). The high quality of the work additionally elevated, with the grades assigned by the evaluators (on a 1-7 scale) rising from 4 to about 4.7. Those who used the textual content editor noticed a slight lower in time spent but additionally a lower within the high quality of the work.

“College-educated professionals performing mid-level professional writing tasks experience substantial increases in productivity when given access to ChatGPT,” the authors concluded. “The generative writing tool increases the output quality of low-ability workers while reducing their time spent, and it allows high-ability workers to maintain their quality standards while becoming significantly faster.”

Other examine findings had been attention-grabbing, too: About 68 % of the employees merely copied and pasted ChatGPT’s output with out enhancing it. Furthermore, regardless that some testers spent extra time enhancing ChatGPT’s responses, the best way the examine was arrange (monitoring output on a minute-by-minute foundation) additionally confirmed that these examined didn’t see any profit in high quality from enhancing ChatGPT’s work—although they spent extra time doing it. That was true even when the examine authors gave incentives for bettering the work.

“Post-treatment, the share of time spent writing a rough draft falls by more than half and the share of time spent editing more than doubles,” the examine discovered.

The examine contributors additionally reported that utilizing ChatGPT considerably elevated their job satisfaction, eliminating the “busy work” that may plague some duties, and had been keen to pay about 0.5 % of their month-to-month wage for a subscription to ChatGPT. (ChatGPT already charges $20 per month for paid entry.) But these examined additionally indicated that ChatGPT didn’t at all times know as a lot as they did.

The different downside, the authors famous—keep in mind, they’re economists—is that employers could begin recognizing the worth of ChatGPT, too. That could put employee jobs in jeopardy.

“At the aggregate level, ChatGPT substantially compresses the productivity distribution, reducing inequality,” Noy and Zhang wrote. “It is also already being used by many workers in their real jobs. The experimental evidence suggests that ChatGPT largely substitutes for worker effort rather than complementing workers’ skills, potentially causing a decrease in demand for workers, with adverse distributional effects as capital owners gain at the expense of workers.”

In different phrases, whereas ChatGPT could assist writers, they could be competing in opposition to AI, too.