OpenAI researchers see current chatbot use supporting — not replacing — workers
The majority of ChatGPT users continue to leverage the chatbot for decision-making support, leading its parent company researchers to see its workforce applications as job enhancements rather than job replacements.
OpenAI’s Chief Economist Ronnie Chatterji noted Thursday during the Semafor World Economy Summit that recent research conducted by the company documented what ChatGPT users actually do with the technology. Conducted in conjunction with researchers from Harvard University, the study found that users turn to ChatGPT for three main purposes: information seeking, guidance and as a writing aid.
“I think what we’re finding right now –– and this is honestly not surprising for a lot of economists who studied technology in the past –– is it’s more of a compliment to workers,” Chatterji said. “It’s more of an accelerant to what you’re doing, rather than a substitute.”
The study found that, given these three prominent use cases, its artificial intelligence chatbot provides economic values mainly through decision-making support. The users fielding these queries are also using ChatGPT in non-work related capacities at a faster growth rate than in professional, work-related capacities.
Chatterji said that these findings hint at artificial intelligence’s future being similar to that of the internet’s: an accelerant to existing careers.
“If you think about the way that most of us… use the internet at work, it doesn’t mean that there won’t be changes to jobs and disruption, but at least for most people using the platform, they’re using it to complement a decision they’re making, look for information, and they’re getting a lot of value out of it,” he told Nextgov/FCW during the event.
Chatterji predicted that firms in the “innovation space” will be developing new products and services that will help gauge the value of incorporating AI into workflows. He added that the research OpenAI conducts on the societal impact and usage of AI draws interest from a diverse range of professionals.
“I think this paper can be useful for elected officials, working government professionals, people at state and local, I think it’ll tell them something about how AI is being used today by consumers,” he said.
The expansion of AI has long sparked fears of human worker displacement, even prior to the advent of generative large language models like ChatGPT, with polling from 2023 noting 75% of Americans believe AI will lead to a reduction in jobs. Chatterji’s work on the subject comes as both private and public sector leaders try to quell concerns about job displacement, with new federal policy aiming to upskill existing employees to work well alongside AI.
Leaders at other AI companies have been more cautious on speculating the level of AI-related worker displacement. In September, Anthropic CEO Dario Amodei said during an Anthropic Futures Forum in Washington, D.C. that conversations around AI-related workforce displacements need to begin so as to mitigate mass job losses.
“If this is something that’s going to affect 300 million Americans and a bunch of people in other countries as well, people deserve to know that there may be these large job displacement effects,” he said.
