April 18, 2023
Dr. Morgan Frank, an assistant professor in the Department of Informatics and Networked Systems, wants to understand how artificial intelligence and machine learning are changing what work will look like in the future. Identifying the recent popularity of tools like ChatGPT and Stable Diffusion, Frank said, “These are all really exciting, but they also all offer the potential to really reshape the way we produce written goods, or visual goods, or even programming and website design. So, this tool and many other tools have the potential to shape the future of work, and they call into question exactly which workers will need to adapt, and what adapting to the future of work really looks like in practice."
"So, the core question in my research group boils down to: how are technology or other labor disputions shaping the future of work and shaping the need for workers to adapt?" said Frank.
Frank’s research uses computational methods to investigate subject matter typically associated with social scientific disciplines. “My background from my master’s and undergraduate education, when I was using tools from complex systems and using big data sets to study expressed well-being, actually gave me a lot of interesting tools to repurpose for studying labor markets as these complex systems, where technology can change small parts of the system, but on aggregate, produce macroscopic changes that probably many people have heard about before. Things like job polarization, for example, boil down to differences in skillsets that tend to get bundled together.”
There has been no shortage of scholarly efforts to understand the effects of such changes. Even so, Frank has demonstrated through his research that there is a lack of consensus around how these technologies will affect workers. Frank said, “There is a lot of work in labor economics on how technology shapes labor. And a lot of the work is very abstract and theoretical. So, there is a consensus that white-collar workers tend to be made more productive by new technology. So, a programmer can do new things with machine learning technology. But if you’re a blue-collar worker, you’re at greater risk of being substituted by new technology. So, the example there would be warehouse workers and robotics. But these are really broad categories and if you force economists to go through lists of specific job titles and say, ‘What about this job? What about that job?’ there are many occupations in which what exactly will happen is a little bit blurry.”
Frank offers two graduate courses in which students can gain exposure to the skills he utilizes to conduct his research. The first course, titled “Network Science and Analysis,” “walks through the history of the network science research community and focuses a lot on real-world examples along the way. Some of these examples will include things like social networks, biological networks, and scientific collaboration networks. And then, towards the end of class, we start to talk about how these tools apply towards epidemiology and economics.”
The other course is titled “Introduction to Computational Social Science.” Frank said, “it is basically a seminar course,” meant “for masters students who are interested in research,” as well as doctoral students. In the class, students “use tools from data science, statistics, and big data to take the insights and theories that come from usually these small social science experiments and scale them up. The goal is to complement traditional social science research with the benefits of big data and the techniques that we can apply to big data.” In the course, “students get the chance to read cutting-edge papers, practice delivering research presentations, as well as hear presentations from scholars in the field, who present their own publications to the class.” Frank said, “I think with master’s students, they’re not aware of the research side and that’s an opportunity. The courses at the graduate level should be taught so that master’s students get a glimpse of the research side and better understand why they’re learning what they’re learning and maybe even discover an interest in research itself.”
--Daniel Beresheim