According to the head of OpenAI, the three industries will look very different in the next few years.
In an episode of the podcast “Learning Without a Teacher”: Olivier GodemanHead of Business Products at the company Creator of ChatGPTshared why he believes three jobs—life sciences, customer service, and computer engineering—are on the cusp of automation.
“I often bet on life sciences and pharmaceutical companies,” he said of his first choice for industries on the brink of disruption due to AI.
Gaudement said the goal of the pharmaceutical companies he works with, like Amgen, is to develop new drugs. There are two components to this, he says: actual research and experimentation, and administration, a labor-intensive process that can be automated.
“From the time you record a prescription for a drug to when that drug hits the market, it takes months, sometimes years,” he said. “It turns out that models are very good at this. They are quite good at aggregation, consolidating tons of structured and unstructured data, detecting various changes in documents.”
Gaudement joined OpenAI in 2023. He previously worked on products for Stripe for eight years.
In the podcast, Gaudement said that while we haven't yet reached the stage where “any white-collar job” can be automated in just a day, he's starting to see strong use cases in areas like programming and customer service.
“Automation has probably not yet reached the level of fully automating the work of a software engineer. but I think we have a clear path to achieve this goal,” he said.
future of software development has become one of the hottest technology debates of the year as AI coding enters the workflow of most companies.
An Indeed study conducted in October found that software engineers, quality assurance engineers, product managers and project managers are the four tech jobs that are cut the most during layoffs and reorganizations.
Finally, Gaudement said functions such as sales and customer service could be automated in the near future.
“I've been working with the guys at T-Mobile, a telecommunications company in the US, to improve the experience for their customers, and we're starting to get pretty good results in terms of quality at a significant scale,” he said. “I believe that in the next year or two we will probably be surprised at the number of tasks that can be reliably automated.”
AI leaders everywhere note that white-collar jobs can be easily automated using large new language models.
In a June podcast, Geoffrey Hinton acknowledged “Godfather of AI“, said that eventually technology will “become better than us in everything” but at the same time, some areas will be safer than others.
“I would say it will be a long time before he becomes as good at physical manipulation,” Hinton said. “So it would be nice to become a plumber.”
“For everyday intellectual work, AI is simply I'm going to replace everyone” said Hinton.
He described paralegals as a high-risk group and said he would be “horrified” if he worked in a call center.
