Delve podcast: Are Digital Tech Workers Coding Themselves Out of Existence? with Emmanuelle Vaast and Alain Pinsonneault (Read Transcript)

Delve podcast, June 16, 2023: Are Digital Tech Workers Coding Themselves Out of Existence? with Emmanuelle Vaast and Alain Pinsonneault
Alain Pinsonneault: Information technology is affecting work, several dimensions of work: it’s creating new jobs, it’s eliminating jobs, it’s profoundly changing existing jobs.
Emmanuelle Vaast: Many occupations are very affected by digital technologies today. What we saw for data scientists, which are very, very deeply affected by digital technology, digital occupation. What we can see for data scientists is going to be seen for many years of occupation. These dynamics of identities, the need to constantly redefine what we do and how different are we from other occupations, are we making ourselves obsolete? These are questions that are also going to be seen by many other occupations. It’s not a question of if it’s going to happen, but when it’s going to happen, and how it will happen.
Robyn Fadden – host: What if just doing your job causes you to lose your job? New technologies have constantly replaced old technologies for hundreds of years, but new digital technologies, namely artificial intelligence and other data-driven technologies, are doing more than replacing old tech—they’re replacing the people who create those technologies in the first place.
Robyn Fadden – host: Welcome to the Delve podcast, an initiative of Delve, the thought leadership platform of the McGill Desautels Faculty of Management. I’m Robyn Fadden, your host for this episode. About 10 years ago, data analytics proved itself a secret weapon to successful, high-growth business. Suddenly, data scientists were in high demand. Universities developed data analytics Masters programs, and people who already had careers in tech flocked back to school to complete degrees that guaranteed a higher income alongside greater value to organizations.
Robyn Fadden – host: Today, data scientists are still distinguished by their advanced data skills and fluency in programming languages. They’re able to generate insights from large data sets and qualitative unstructured data. Yet their skills must constantly be updated to keep up with the rapid evolution of digital technology—and how that technology has changed organizations they work for. It’s a shift in occupational identity that has changed the entire outlook of data scientists and others who work with digital technologies, such as social media managers, investment analysts, and software programmers. How can they navigate these changes and still find meaning in work that might be done by AI in the near future?
Robyn Fadden – host: New research by Alain Pinsonneault, a Desautels Professor of Information Systems and IMASCO Chair in Information Technology, and fellow Desautels Professor of Information Systems Emmanuelle Vaast examines how digital technology enables and threatens occupational identity—and how data scientists cope with the associated tensions. In their research, they quote a data scientist who says that, “The role of the data scientist is to be the value creator—the bridge between statistician/computer engineer/etc. and key decision makers. … That is how I reconcile the blending of disparate skill sets (statistics, programming, business acumen) into the ideal role of data scientist.” We know that occupational identity changes over time as social and cultural contexts change—and as new technologies are introduced. But clearly, occupational identity is complicated in the case of data scientists. I asked Emma and Alain about how occupational identity is formed and why stability in occupational identity matters.
Emmanuelle Vaast: Occupational identity is the idea of how people are able to define themselves with regard to their work. So usually, we think about occupational identity as the answer to the question of who are we and what do we do with regards to our work. And it’s very important to be able to have an occupational identity that’s fairly clear. What is a journalist, what is a data scientist, what is a professor, what is an accountant? We know what these occupations are in what they do. And being able to have a fairly clear occupational identity, it is important for people to be able to give meaning to their work. Usually occupational identity is defined over time as more and more people are engaged in a particular line of work. But the idea is that occupational identity is relatively stable for now. So that people are expecting to know what this occupation is.
Robyn Fadden – host: Being able to give meaning to work and have a sense of stability in it affects people’s broader sense of self. Why does occupational identity change and how does that affect people’s everyday work?
Emmanuelle Vaast: Occupational identity can change over time in the long run, because contexts change, because new technologies, including digital technologies come over time. And so what people do is going to be changing over time. So usually it can be stable for now, but it’s going to be changing. Usually, what we think about occupational identity is that this change is going to be relatively slow. And then it’s going to stop over time. And what we show in our study, is that actually, there may not be any more this idea of stable-for-now occupational identity, maybe it’s a constant shift in occupational identity are what is happening now. And will continue to happen in the future.
Alain Pinsonneault: So basically, the now part of “stable for now” is shrinking dramatically.
Robyn Fadden – host: That can’t be easy for people to navigate, especially at the rate technology evolves and is affecting occupations. What led you to notice these changes to occupational identity in digital tech occupations in particular?
Emmanuelle Vaast: It was more than 10 years ago, many, many students talked to me about wanting to be a data scientist, many people talked to me about data scientists, the Harvard Business Review, or the very famous article about data scientists being the sexiest job of the 21st century. And I was really, really interested in data scientists, without really understanding what they were and what they did. So to get an occupational identity. And I decided to try to see what these data scientists were doing and who they were. And to do so I started to follow the past and the discussions on an online community called Data Science Central. And this, it’s a very, it’s an old, for the occupation. It’s an old online community. It was composed of many data scientists. And what was really surprising was that data scientists themselves didn’t seem to really know who they were and what they did. So the question that I was myself asking as, as a non-data scientist, were questions that many data scientists themselves asked. So this really made us made both of us very curious about the occupational identity of data scientists. So we decided to dig into this. And we realized that digital technologies were really at the core of many things, of many of the tensions related to occupational identity that data scientists experienced.
Robyn Fadden – host: Tension at work, like around projects or even people, is difficult to navigate and can make just being at work difficult too, obviously. But tension around the purpose of your role and the work you do goes even deeper and can sow self-doubt or resentment. Could you talk more about these tensions and why they’re currently specific to occupational identity in digital technologies?
Alain Pinsonneault: It rests on fundamental characteristics of digital technology. Digital technology makes the boundaries between occupations fluid, and it’s made change happen very fast. It rests on two fundamental characteristics of information technology or digital technology, the first one is often called Convergence. And it’s the ability of information technology to integrate things, people, objects that were separated in the past that were not connected. And it integrates them in a way that changes occupation. If we think of technology, for example, Internet of Things, it’s connecting things that were previously not connected. AI is a good example of that. Data analytics is a good example of connecting data that that were not connected in the past. And this convergence brings an interesting tension of similarity and distinctiveness among and between occupations. It blurs the line between what is an occupation and what is not an occupation.
Robyn Fadden – host: We’ve seen this in journalism too as the web grew and changed, as we moved away from print, as the media landscape changed incredibly: who’s a journalist and who’s a blogger or a podcaster? There’s a difference but the overlap between them has grown over the years, and some journalists have become bloggers and vice versa.
Alain Pinsonneault: And it changes how journalists see themselves and how they define themselves. They don’t have to describe things as much anymore because it’s already done by blogger, but they need to explain things. So the notion of occupation changes. What is the difference between a data scientist and a statistician in software engineering? Same thing but there’s an overlap on who they are. What is the difference between the librarian and Googler? Also there’s an overlap. This convergence brings this tension of similarity and distinctiveness among occupation. The second key characteristic of digital technology is the idea of Generativity, which is that technology constantly evolves, and quite rapidly. So it brings attention to obsolescence and how the jobs and occupations stay, and persistence and obsolescence. And that’s true for all digital technology and data scientists, especially.
Robyn Fadden – host: Naturally, these characteristics of convergence and generativity in information/digital technology affect occupations in digital technology. What does that look like though, how does digital technology make these occupations distinct from other occupations, whether in non-digital technology or otherwise?
Alain Pinsonneault: Digital technology occupation is a bit different because technology plays two fundamental roles. First, it’s a tool that helps data scientists or people working with digital technology—it’s the tool that helps them do their work. They fundamentally rely on digital technology to do what they’re supposed to do. The second characteristic is digital technology is at the core of their very existence. So it’s a tool, but it’s also what they create. Data scientists, for example, partly enable AI or data analytics technique. It’s the core of their job. And it’s quite ironic that these tools might actually replace data scientists because it’s automating their jobs. So they’re working with the tool to create the tool that might replace them as an occupation.
Emmanuelle Vaast: Yes, it’s very as Alain said, it’s very ironic. And data scientists were very aware, the data scientists that we studied were very aware of this, and they were very often talking about how they were developing these new tools, these AI technologies in particular, and how they were making themselves obsolete, while doing so, and we call this persistent extinction. So, the idea that the occupation in particular the occupation of data scientist is still there is there for now, their job openings are not decreasing. But at the same time, maybe in the future, it will be and partly because of what data scientists do, but I think also, many occupations are very affected by digital technologies today. What we saw for data scientists, which are very, very deeply affected by digital technology, so digital technology occupation, but what we can see for data scientists is going to be seen for many years of occupation. These dynamics of identities need to constantly redefine what we do. How different are we from other occupations. And are we making ourselves obsolete? These are questions that are also going to be seen by many other occupations. So it’s not a question of if it’s going to happen, but when it’s going to happen, and how it will happen.
Robyn Fadden – host: There are many AI tools being created for so many different fields, and they’re solving data management problems, they’re adding incredible analytical power to organizations, and they’re also changing and eliminating jobs. Take self-checkouts or even chatbots, for instance. Is this obsolescence inevitable? Or is there typically a chance for occupations involving digital technologies to change?
Alain Pinsonneault: I think there’s a chance. I think any technology brings, and digital technology brings these tensions that I mentioned, and the tension of potentially making jobs and occupation obsolete. But I think it also there’s a threat and opportunity in any technology, and the idea is to be able to define and redefine their own occupation, their jobs, so that technology is an opportunity for them, it’s a different way of doing their job and of doing what they’re there to do, so adding value. So I think the fact that it’s changing is good, in a way. Is it a bug or a feature? We need to think of it as a feature rather than a bug. It’s an opportunity rather than a threat. But it needs to be actively managed.
Robyn Fadden – host: Looking specifically at data scientists, as you did in your research paper, who did you talk with? Where did you gather your information from? And what methods did you use in your research process?
Emmanuelle Vaast: This is a qualitative study. What we did was we both did some interviews of data scientists, but mostly we collected some data from Data Science Central itself. So the interviews were interviews of data scientists, among which were members of the online community of Data Science Central. So these, these interviews were useful to get a sense of data scientists were and how they, how they talked where they considered to be important, and also to check that what we thought we observed in in the online community, were things that data scientists actually also experienced, we wanted to make sure that it was this was also observed by them and experienced by them. But mostly is a bulk of the data were the first six years of entries from data science Central. So all of the blog posts, all of the discussions that took place for six years in data science Central, and this was really a nice data set, because it enabled us to observe change over time. So the evolution in, in the occupations evolution in, in the technologies that it used, and in the ways in which on in which the online community members actually defined themselves. And from this six-year data set, we were able to identify what we called episodes in which the identity works, or the ways in which these data scientists struggled with the findings themselves were particularly relevant.
Robyn Fadden – host: In looking at all that data and all those people’s experiences, what did you find out and how can these findings be applied?
Emmanuelle Vaast: One of the key findings is this idea that digital technologies make it that occupational identity cannot be considered stable for now anymore. So it’s going to be changing over time. We have this push and pull with digital technology where it can be both an enabler of the occupation and its identity but also a threat to the occupation and its identity. It’s really important to try and understand how this push and pull is actually happening, and how people, the members of this occupation, are able to solve the tensions associated with your occupational identity.
Alain Pinsonneault: This has important implications for research for any scholar who studies technology and digital technology. Because these tensions that we observed and the way data scientists manage this tension, the tensions are present in almost any digital technology implementation of any technology. And if we think about normal or regular technology being implemented in the workplace, it always constitutes both a threat and an opportunity. There’s tension there of what the technology does, what the human does, and how to cope with that. If we think about AI, for example, AI is implemented in numerous contexts now. And it challenges one fundamental principle of management, which is if you’re responsible for something, you ought to be in control of that thing. And what AI does, fundamentally is it associates, it takes over the control of some part of the job and some part of what is done. But in the end, the human is always responsible for the decision and for what happens. So it dissociates, it creates a huge tension between what you control what you don’t control, while being responsible for the whole job. And there is no decision that’s made. So there’s a tension there. What we found in terms of how data scientists deal with this tension can be used and provide some insights into these other domains as well.
Robyn Fadden -host: As far as technology goes, there’s also a tension there between the technology leading the occupation and the occupation leading the technology. We see that in AI especially. How do people navigate when it might seem the technology is taking over from them?
Alain Pinsonneault: The first thing I think it’s understanding technology, understanding what the technology does, what it doesn’t do, and what it does well, looking at the job from a holistic point of view, our own occupation, our own work from a holistic point of view, and understanding how technology fits in that. And allowing technology to do what it does best, creating some slack and allowing the individual to focus on something else that adds value. The other thing is staying nimble and always on – watch for things that are changing the way we set this table, for now, the NA part shrinks dramatically, and it’s constantly evolving. So you constantly have to rethink how new technology will come and how that will redefine your job and how you need to redefine your job. That’s what we draw in terms of practical application here.
Robyn Fadden – host: In your research, did you find that people who expressed that their jobs had less meaning or value felt that this made their jobs more difficult?
Alain Pinsonneault: I think it varies from that research from other research also that I’ve done on the impact of it on work. When people understand the technology well, and when they’re proactive in taking advantage of that technology, it’s actually make sometimes make their job more easy and more useful and more challenging, but also potential of adding more value. And when you feel it’s a threat, and you’re reactive, and you’re passive and you’re resisting, then it obviously makes your job challenging, quite challenging.
Emmanuelle Vaast: Yes. And I would very briefly add, also, I think, many people like data scientists who are again, you know, “sexiest job of the 21st century,” almost at the top of the hierarchies of jobs these days, feel very strongly this sense of ambiguity. So on the one hand, they are very, they’re celebrated, for the expertise for the ability to develop new technologies and to do great things with data. And on the other hand, they are also keenly aware that what they do that makes them great today might be the very source of what makes them not so great in the future. So the sense of ambiguity means that they are very good at their work. But there’s also this sense of, for now, for now, I’m good at my work, but in the future, it will need to change.
Robyn Fadden – host: What can be learned through your research about the investigation of occupational identity itself, as both a concept and as something personal to individuals?
Emmanuelle Vaast: Occupational identity is one of the main ways to have meaning associated to the work, so if occupational identity is under threat, or if it’s not distinguished enough, from other occupations, it’s really difficult for people to be able to bring meaningful value to their work. So being able to engage in identity work and redefine, and as Alan said, be nimble with regard to who we are in what we do is very important. And in this regard, I think one of the points this study is making is online communities can be a very good locus for discussions around occupational identity. So usually, when we think about online communities, we think about these online forums where people can talk about their work and can find new ways of doing things and be better, have best practices. But what we found was that it’s not just that. Of course, best practices are important and useful, but online communities can also be a way for people and a place for people to actually exchange about their work when they lack such places to do so in other ways.
Alain Pinsonneault: And that’s so important, research clearly shows that people having a meaning in their work and value in what they do is fundamental to motivation, to performing, to enjoying the work. That’s fundamental.
Robyn Fadden – host: So even if AI changes the work landscape and work doesn’t look like it used to, people will continue to contend with these fundamental issues, people will still look for meaning or value in what they do. It’s still important to people to contribute something meaningful to the organization they’re working for or, even more broadly, it’s important to contribute something meaningful to the society we all live in.
Emmanuelle Vaast: It really is. And there’s also this the other way around, which is society provide a particular mirror to their identity. And so members of occupations internalize this and, in a way, this becomes part of their identity struggles and shifts of identity.
Robyn Fadden – host: As digital technologies continue their rapid evolution, predictions can be made about how they’ll affect the future of work—but listening to the people who spend time in and around these occupations today reveals many tangible effects of digital tech’s evolution. Since organizations rely on both data and workers, they might want to listen too.
Robyn Fadden – host: Our guests today on the Delve podcast were Desautels Professors Emmanuelle Vaast and Alain Pinsoneault, discussing their research on occupational identity among data scientists and others who work in new digital technologies. Find out more about their research and find out more about Delve on
Robyn Fadden – host: Thank you for listening to the Delve podcast, produced by Delve, the thought leadership platform of the Desautels Faculty of Management at McGill University. You can follow DelveMcGill on Facebook, LinkedIn, Twitter and Instagram. Subscribe to the DelveMcGill podcast on your favourite podcasting app. And subscribe to Delve’s email list by going to