Artificial intelligence is beginning to change the entry-level job market. AI often automates the repetitive tasks that might otherwise allow new employees to learn on the job, and some worry it could eventually replace those jobs altogether. A study from Stanford University found these changes were most likely to affect “AI-exposed occupations” such as software development and customer service.
Lucas Hellberg is an enterprise reporter for the Daily Emerald newspaper at the University of Oregon and an elections reporting intern for Lookout Eugene-Springfield. He recently wrote about how AI is changing entry-level jobs in Oregon and joins us with more details.
Note: The following transcript was transcribed digitally and validated for accuracy, readability and formatting by an OPB volunteer.
Dave Miller: From the Gert Boyle Studio at OPB, this is Think Out Loud. I’m Dave Miller. Artificial intelligence is changing the market for entry-level jobs. AI often automates the repetitive tasks that might otherwise allow new employees to learn on the job. Some people worry it could eventually replace many of those jobs altogether. A study from Stanford University found that so far, at least, these changes have been most likely to affect occupations in software development and customer service.
Lucas Hellberg recently wrote about how AI is changing entry-level jobs in Oregon. He is an enterprise reporter for the Daily Emerald at the University of Oregon, and an intern for Lookout Eugene-Springfield. He joins us now. Welcome to the show.
Lucas Hellberg: Thank you for having me.
Miller: You started your recent article with some quotes from a UO computer science student who had a software engineering internship at Microsoft last summer. What did he tell you about his experiences at that internship?
Hellberg: He told me that AI is having a big impact on software engineering specifically, and he told me that he was told to use AI before he asked a human for help. He said it was like his mentor.
Miller: Did he feel like he learned from his AI mentor?
Hellberg: It sounded like he definitely did. It helped him learn a new programming language that he didn’t necessarily learn before he got to that internship.
Miller: So that’s just one student. How did his experiences fit into the larger picture that you were able to put together?
Hellberg: His experience showed how AI is having an impact on how we’re training younger workers at institutions, as well as at companies. It showed that AI is having an impact in software engineering, but it also opened my eyes to look at other fields like investment banking, where it’s having less of an impact, but it’s also speeding up a lot of the tedious tasks, particularly with data analysis.
Miller: You pointed to a fascinating recent study that I mentioned briefly, but we can dig into it more now, this came out of Stanford. It found a 6% drop in younger employees in a number of sectors in the last few years, but an 8% increase in slightly older workers, in workers ages 35 to 49. How did the researchers explain that?
Hellberg: The context behind that is essentially that a lot of companies, particularly in software engineering, but other fields like marketing – which is a little bit less AI exposed – they want the older workers to train the AI because they’re going to be better at training it, because they have the experience with everything and they know how to catch the mistakes AI makes, which younger workers don’t necessarily have the skills to do. In some cases, it’ll mask their experience. They could produce something that someone who’s like 10 or 20 years into their career could produce, but not necessarily something that somebody who’s an early career professional.
Miller: What are examples of the kinds of tasks that people in various jobs would have been forced to do in the past that AI is increasingly likely to be doing now? And what did you hear from experts about what it means that people in those entry-level jobs will no longer be doing those tasks themselves?
Hellberg: Yeah, entry-level workers in many fields, particularly in software engineering, they gain through repeated exposure to problems, coding and other things like that. But now a lot of that code is being generated by AI. So what that means for people in software engineering, there’s less ability for them to learn from repeated exposure, because they’re doing a lot of different things now that the AI is making all the work for them. And at companies like Intel, it means that you can make a project with a smaller team. So it’s also raising expectations for early career professionals, but also giving them less exposure to repeated tasks, which is what they learn from.
And there’s a concern that there’s exposure to this, not just in software engineering but other fields like maybe even investment banking, that it’s going to be harder for companies to build these senior level workers because they’re not trained as well, potentially. And that’s something that universities are adapting to.
Miller: So if entry-level employees are increasingly likely, now and in the years to come, to be managing AI systems as opposed to doing their own software engineering, data entry, draft writing or data analysis, what might they be missing out on in terms of the holistic skills required to do their jobs at higher levels?
Hellberg: There hasn’t been a lot of research showing what that is happening, because generative AI just came out about three years ago, so it’s still a pretty new thing. But there is definitely concern about that. One thing that I’ve learned is that when you look at specifically what you were talking about, the workers need to get their soft skills, like critical thinking, seeing the bigger picture, and communication skills. And that’s harder to do now with AI because AI is doing a lot of that, but it’s not necessarily doing the best job of it, from what workforce experts have told me. And that’s something that colleges are emphasizing or trying to emphasize more, but it’s harder because you’ve got to get to the curriculum and those aren’t necessarily in the curriculum,
Miller: So how are college or university instructors that you talk to thinking about these issues?
Hellberg: At Lane Community College, for their software development degree, they’re getting rid of some of the coding classes and they’re replacing them with software design classes. The reason they’re doing this is because they believe the people in software design, they want to teach them higher level skills because they believe those skills are going to be the most important, like the architects, the complicated things that AI can’t do, and it’s going to take a while for it to do correctly.
And then at the University of Oregon, they’re not necessarily changing courses. They’ve added some courses, but they’re not changing the curriculum as a whole for their computer science program. They’re redesigning assignments to stop AI from short circuiting learning, and they’ve also introduced a new entrepreneurship class where they work with companies like Intel and they build things for them. And it’s something that they’ve created to help students get more practical experience that’s becoming increasingly important.
And the goal of the program, they say, is to make sure students have that practical experience that they would get an internship, which is becoming increasingly harder to find in the tech sector.
Miller: One of the most striking things about the changes that AI is ushering in is the speed of these changes. If we had talked certainly two years ago, but maybe even just a year ago, I think that even then the conversation would have been different. How are the people that you interviewed thinking simply about the pace of this change?
Hellberg: Things are definitely moving faster than people expected. I was talking to somebody at the Springfield Chamber of Commerce, and she was saying that even for people in their 30s and 40s who are on the younger side, they’re having trouble getting a hold of AI, and they’ve started having workshops to teach them these skills. It’s something that everyone needs to adjust to, and even young people are having trouble keeping up with it. Because they’re not necessarily teaching people in school how to use AI. Some classes are, I’ve definitely had some classes myself where they’re teaching us how to use AI to do certain things outside of journalism.
It’s hard to adapt and it’s definitely moving quickly. But people aren’t yet scared that AI is going to replace their jobs – at least not yet.
Miller: Do you feel the same way? I’m curious as somebody who is going to graduate soon, how did everything you reported on make you think about your own future?
Hellberg: Well, it made me see that the job market is definitely getting more competitive every year. It’s more competitive than it was three years ago, we can see that in the data from the Oregon Employment Department. There’s fewer openings and more things require practical experience. So I don’t see AI replacing my job. I just see it getting more competitive and I see the need to differentiate myself from the competition and sort of stand out. I think building a portfolio, and just having the soft skills like communication, writing, seeing the bigger picture of the skills that are going to matter increasingly.
Miller: Lucas, thanks very much.
Hellberg: Thank you for having me.
Miller: Lucas Hellberg is an enterprise reporter for the Daily Emerald and an elections reporting intern for Lookout Eugene-Springfield. He wrote recently about the way in which AI is affecting entry-level jobs.
“Think Out Loud®” broadcasts live at noon every day and rebroadcasts at 8 p.m.
If you’d like to comment on any of the topics in this show or suggest a topic of your own, please get in touch with us on Facebook, send an email to thinkoutloud@opb.org, or you can leave a voicemail for us at 503-293-1983.
