#212 The History of Data and AI, and Where It's Headed with Cristina Alaimo, Assistant Professor at Luiss Guido Carli University

Hi, I'm Maggie, VP of Curriculum at datacamp, and I know firsthand how overwhelming it can be to keep up with the rapid changes in data and AI. But staying ahead doesn't have to be hard. That's why our curriculum team has curated the best learning content to help you and your team stay ahead, covering everything from data and AI literacy for busy executives to cutting edge generative AI for developers, and everything in between. Whether you're upskilling yourself or your team, we make learning practical, engaging and flexible. It just works. Power your data and AI transformation journey on datacamp.com.

This idea that how we organize things collectively has a very close link to what kind of data we produce, how do we use them, how do we mobilize them? So we start with these clay tokens, this little sort of artifact made of clay that had small some shapes. They were like cylinder shaped or different shape. They were stand for something, for instance a ship or a day of work. And they were exchanged in the Mediterranean as a symbol, a token of a transaction.

Hello everyone. I'm Adele Data, evangelist and educator at datacamp. And if you're new here, Data Framed is a weekly podcast in which we explore how individuals and organizations can succeed with data and AI. One thing we like to do on dataframes is cover the current state of data and AI and how it will change in the future. But sometimes to really understand the present and the future, we need to look into the past. We need to understand just exactly how data became so foundational to modern society and organizations, and how previous paradigm shifts can help inform us about future one, and how data and AI became powerful social forces within our lives. Enter Christina Alaimo. Cristina is an Assistant professor of Research of Digital Economy and Society at Luise University in Rome. She co wrote the book Data Reinventing the Market Economy with Yanis Kalinikos, professor of Organizational Studies and the Cisco Chair in Digital Transformation and Data Driven Innovation at Luise University. The book offers a fascinating examination of the history and sociology of data and throughout the episode the we spoke about many of the themes covered in the book, from the first instance of where data was used, to how it became central for how organizations operate, to how usage of data introduced paradigm shifts in organizational structure and culture, and a lot more. If you enjoyed the episode and the Data Frame podcast, make sure to rate it wherever you get your podcast. And now on today's episode.

Cristina Alaimo, it's great to have you on.

Hey, hello. Thanks for inviting me. It's great to be here.

Yeah, I'm Very excited to have a chat. So we're going to be talking about your book today, Data Rules. Very excited to talk to you about it, but maybe to set the stage for our chat and our conversation, walk us through the motivation behind why you and Yanis Kalinikos, your co author, decided to write Data Rules.

Well, you know, your show is all about that. We live in the age of data. So we really, we cannot basically do anything any longer without data. We do stuff through data, by data and with data. So we do groceries. You know, we chat, we record shows, we study, we work through data. So our everyday life has changed because of the presence and involvement of data and of course, database technology. But these little tokens, these little artifacts that circulate and make things possible differently from how things were made possible before. Of course, as our everyday life changes and there are sort of reshuffling, I would say, of how things are done. So perhaps some things become similar, like, you know, studying and making and buying grocery becomes increasingly similar as we use data and this digital device to do that. Also, our institution changes, so our institution change, our organization change. And so this double level, I think the level of everyday life and the macro level, the level of how institution also respond to this big shift. We felt there was something that was worthy, let's say, to go a little bit deep, exiting from a sort of, let's say, narrow view of data only as input to large system, large model, but data on their own as artifacts of human making and instrument of communication and tokens through which knowledge is transmitted, through which we see the world, ourselves and the world, and we do things. So that's a little bit this, this big reshuffling of everyday practices, but also restructuring of institution, I would say, are central concerns of the book.

And in a lot of ways, the book does a very thorough job at looking at data through a bit of a sociological lens and a historical lens and really tries to map how data has impacted, as you mentioned, institutions, but how we communicate, how we work together and how we really think about the world today in the 21st century. I want to maybe first start at the history side of things. You actually do a pretty thorough job at mapping out the history of data from how we used to record information on slabs of stone to today's applications and digital technologies and the digital exhaust that are produced by database technologies. But maybe walk us through the history of data in a bit more depth and I would love to see what you think are pivotal moments and milestones in the history of data that has led to the paradigm shift that we're in today.

That's great because we sort of look back in order to look forward. I mean, our approach to history is a very sort of instrumental approach. We want to learn from the past in order to learn how to ask the right question for what's going on today. So that's very important because of course we are not historians, but we want to learn from history. And we do so by retracing, as you mentioned, pivotal moment, I would say, in the role of data within society and the links between data like social cognition. How do we understand the world and institution. This idea that how we organize things collectively has a very close link to what kind of data we produced. How do we use them, how do we mobilize them? So we start with these clay tokens, these little sort of artifacts made of clay that had some shapes like they were like cylinder shaped or different shape. They were stand for something, for instance, a ship or a day of work. And the way it changed in the Mediterranean as a symbol, a token of a transaction. And that was the first moment in which that we know, of course, in which as transaction exchange were recorded, they ceased to be only something that was happening in a given moment in time and in a given place. And they became something that could be extracted from that moment. Thought as a social event, as a sort of objectified event upon which something else could be built. Practices of taxation. Because I could record the exchange between me and you and I could encode it into a token. And this token could be stored then in a year time I could recount how many transactions we made and I could extract like a tax or something else. And of course this was used exactly in proto bureaucracies at the time. They were the temple. And effectively taxation was born almost by that. And I would say the most important thing was born by the manipulability of concept. Very complex concepts such as exchanged. They were made objective, so to say, with a life on their own, because they could be encoded into data. The links between data and the making or the emergence of institution are also recounted in another episode where we talk about the rise of statistics. You know, the numeracy, which really signed a very important moment in history. Because to statistic to the quantification of collective behavior is linked government and how do we govern such behavior. And therefore the making, if you want, of modern institution, of modern public institution, the state. That's very much linked to statistics, which is the science almost of the state. But this is also linked to data. So how do we gather data? How do we make those data the representation of some things that we start looking at a product of society that exists only insofar I can count these events. So it's really sort of a very interesting feedback loop here. I do measure society through the counting of some event. And for this reason I almost make society as something representable. And therefore this, I would say, is a second important moment, a third moment, and I go to us here is the making of corporation. The emergence of modern corporation that the business historian Alfred Chandlers recount as the change when these modern corporation had to use internal data in order to keep track of their goods, their resources. Think about the beginning of last century or the end actually. So the end of the 19th century. Think about modern railways. You know, think about all this technology coming and this big corporation, the organization expanding, changing places, changing, also diversifying, also their operation, having to hire people. How do you suddenly track all these different goods and workers and way of working? So you need to start recording internal data. So that's the idea of Chandler. And as you keep on recording what the people that work for you do, how do they do that? Where do you locate goods, physical resources? Then you become a corporation. So you become a different kind of organization that can expand in scale and scope and give rise to modern ways of doing business. And that's also very, very interesting because once again, from history we learn that there is a link between institutions, data, work practices. Chandra, for instance, links the emergence of managers of clients to the use of data. The administrative science, you know, is the science of data. And yes, I think that these three moments are very important. And then of course, perhaps the last one which is the most perhaps relevant for us is the advent of machines and how they change basically everything within organization. So first the adventure machine, the Hollerith punch cards, and then the advent of computers as digital data makers. And this is something that really first within organization and thereafter for us, for society and large as such has been a pivotal moment. Data ceased to be just linked to specific practices like archival data, to the archivist or accountant data to accountants. With digital technology, they become almost universal token where different things could be recorded, represented, manipulated, and also conjoined, you know, like combined. And new things can be learned from this recombination of data that up until that point were belonging to different areas of work, of life. So yeah, really like fast walk through thousands of years. Yeah, I mean it's so fascinating. I can talk for hours. You need to stop me here and tell me enough let's talk about something else.

What's interesting about what you're mentioning here is that you try to see kind of these four pivotal moments, three pivotal moments, the history of data. And you're talking about kind of the age of digital technologies at the moment that we're living in, where the production of data no longer, correct me if I'm wrong here, no longer serves a purpose where you're doing it just for a use case, such as tracking your resources. But now you use your phone, that data of how you're using your phone is being logged, sent to the cloud. And in a lot of ways, I think that the current paradigm that we're living in or the current era that we're living in is also ushering in a new category of use cases or a new category of intelligent machines. Especially when we look at kind of intelligence, generative AI and how many of the AI technologies that we use today are based on a lot of the data that we've been producing. So maybe given this perspective, right. How do you look at historical perspectives on data? How can it inform what we're seeing today with generative AI as unlocking a new way of interacting with digital technologies? Because data is enabling intelligent machines that can mimic human, human behavior, like human text and you know, make intelligent, semi intelligent sounding text.

Yeah, definitely. This is, I would say perhaps someone in some year will look at this as another pivotal moment in history. So definitely there is something going on here which if we look at it from a data lens, because that's also very important. Current discussion on the role of data and AI or AI, they always look very narrowly or, or well, most of the time, let's say a data's input to these machines, like huge data set or different data and this intelligence machine that learn from these different data and can churn it and chunk it and produce something resounding as intelligent. But if we adopt a slightly different perspective, which is what we suggest in the book, and we look at data not just as inputs to this big device is, but to artifacts of human making. Human means here, societal, doesn't mean that someone goes there and do it, of course. So with their links to culture, with their links to organization, data are produced by some organization, for some scope, for some aim, with their links to way of doing things, and we start to unpack little by little all of this different aspect of data, then we can ask the right question to the AI current wave, or at least let's say a broader range of questions, not just the question of how intelligent These machines are. But for instance, what kind of intelligence do they propose? Or what kind of intelligence do they propose? And how does intelligence infiltrate in the existing intelligence of institution of organization? What's the change of paradigm here in knowing things and what kind of consequence does it have from the way we do things? Because if we go back and we recall what we just said about the history of data and their links with way of working, the emergence of institution, then we look at this novel data churning machine, let's say, and we need to ask, so what kind of institution will they facilitate the emergence of? Or do the existing institution we have? So the existing kind of organization, are they still fit to deal with this huge knowledge change that these large language model, these, you know, artificial intelligence proposes at advance? So I think that's my answer to your question, is like, let's open up the discussion of data, avoiding these narrow routes of data as input and opening up data as artifact, and then start asking what's the role and what's the big changes here? There is another, more technical aspect too of course here which is the process of datification, the process of working with data changes with this machine they propose entirely new data practice, entirely new way of manipulating data, entirely new way of accessing data and tracking these data work, accessing this data work. What does that mean? So again, if we look at data work not just as sort of a technological process, but more as a social, sociological also process. These machines are in society, they, they are part of our society. And then there is a constant feedback between society and technology. Then you know, also this is interesting. So engaging in bit more detail with the kind of changes in data work, the operation, the data operation, this machine do and ask questions, question around the practice connected to it, the kind of cognition, social cognition linked to it, the professions that will emerge will just cease to be relevant because of the novel kind of data work archive like we learn from history. At some point data were not any longer scribbled on paper. And all of these archivists that were working for organization, they were no longer working for organization because there was no need. But a new class of clerks emerged, manager, A new class of manager emerged. So you see, it makes many of the questions we ask ourselves today, will we lose jobs a little bit more complex and perhaps are a little bit more grounded in the complexity of our society.

I would love to pick your brain on how you think exactly or how you predict society can change or evolve based on this current paradigm of AI that we're living in, but you mentioned Something called deification here, which is actually a very central concept of the book. Maybe first to set the stage, what is deatification?

Well, as simple as I can, I think is the translation, the encoding of events or episode of life into data. So is this kind of idea of representing, recording or representing something of life into data, tokens into data. So that's I think the simplest way I can go in order to. That's of course that's just the beginning, but I think, yeah, that's a basic idea of it.

And you mentioned there's a bit of an ecosystem when it comes to deification that you expand upon in the book from a sociological perspective, how organizations approach it. Maybe expand on the observations that you've had from the book while studying this ecosystem of datafication. I'd love to understand here a bit more.

So yeah, because of course this recording is not a simple recording. It's not as simple as it sort of may sound. And this is very much linked to the fact that data are not just input or, but they are the product themselves of a very complex process that has several stages, a life cycle and several actors involved. That's the ecosystem you mentioned, many different actors, but also many different links to our social reality and economy. So perhaps an example will work better here in order to trace the boundaries or the characteristics of this ecosystem, the programmatic advertising ecosystem. So one of the example we use in our book and the programmatic advertising is the automated exchange, buying and selling of advertising space online. This is done through auctions, auctions that last very, very little, like 0.03 milliseconds. And they happen every time we open a newspaper online. There are several ad space that needs to be assigned this ad space trigger data object, a very complex system. This data object embeds some specific data about who is opening the page, the price of the ad, blah, blah, blah, and thousands of platform bids for this space. The auction happened. Ads are assigned to space and I see my newspaper with my well placed ads. Now all of this, it's an immense complexity of stuff that happened all through data and data objects that coordinate the exchange. But the interesting bits here is what is it that is actually sold there? Now that's the interesting bit. That's the social sort of science of data or asking questions. Because marketing always and advertising always had its problem of measuring the attention of users and therefore putting a price on a space for an ad. And at some point there was the conviction, the idea that this could be solved via data and digital technology. As we have more Data about user and user behavior. And therefore we can be more precise. But there is a lot of work still going into defining what is it exactly that we can measure and how can we put a price on it. So for instance, we can measure click. There is not yet any way of going beyond the screen and really looking at what I'm doing. So I need to use proxies. I can measure, for instance, the level of the screen eye level. I can measure the click. I can measure impressions that are defined as the likelihood of users seeing an ad. All of this gives rise and depends upon a huge ecosystem of companies. Actors that measure, that invent proxies, that invent ways of verifying that these measures are reliable, that give badges to the reliability that enter in getting more data in order to have more elements to increase the likelihood that exchange those data. So this is what we mean by ecosystem and the ecosystem of dertification, not just actors that exchange data, but actors that are almost created out of data, needs the need of verifying what an impression is, certifying what an impression is. And if you think about how this is linked to existing practices, for instance, an impression has not been invented by digital data, but was already existing and now digital data change it. So I think that this idea of dedication and data ecosystem is an idea that allow us to to see that making data. So the tatification is not a straightforward process. So it's always a process that involves decision making, links to how things were made before and how can we translate it now that we have. Digital data involves different actors, regulation and governance, and increasingly so involves the note straightforward thinking of what is it that we sort of select in order to be recorded into data and for what purpose and how that works for us. And then as you multiply this question across all the actors of the ecosystem, you can see that this is an immense complexity because each actor will have its own idea and then there is technologies and layers of technologies. So yeah, that's very interesting because it allows us really to sort of disentangle a little bit this monolithic idea of data and instead articulate it across actors practices. And I would say also time.

Yeah, that's really interesting. And actually this description of the digital ad ecosystem, if I connect it back to our earlier discussion on the advent of large language models, I would anticipate that the introduction of large language models into this ecosystem would completely change the nature of the ecosystem. Right. So let's say instead of measuring clicks and measuring impressions, a chatbot asks you a question and you're able to answer it and the system is able to react in real time based on the sentiment that you have in your answer. Maybe the voice and tone of your voice. If you're actually recording an audio instead of answering in text. How do you anticipate potentially the advent of data producing AIs or AIs that are able to understand these nuances on an individual basis to impact or change these systems as you look at them or these ecosystems?

Perhaps I can say a couple of things. The first is that no matter what kind of technology and sophistication this technology has, there is always a decision within made somewhere and repeatedly so on. What is it that I want to select in order to define a specific sentiment or a specific reaction in user? This is not straightforward. So that's very important. We should not fall into the trap of thinking that as technologies evolve therefore it becomes, let's say objective. No, there is always a set of decisions that have to be made, but there is always a decision of selecting this other than that and making it your data, on top of which you do then your work. So that's very relevant. However, and you were absolutely correct in saying so as of course there are shifts in technology and the use of data and the links that can be made among different data. Of course all this ecosystem of production also change and it's unavoidable because you know, new actors will emerge that will be better placed using these technologies or packaging these technologies into apps. Let's not forget that this is what we are looking at right now. We are living in the sort of ampification I don't know of like you know, these intelligence technologies. And that's. We still is slightly different from what we have seen so far and there are some commonalities, but it's still something that we are seeing now at a very larger scale. And there are other ways in which and we were mentioning before data will be worked and used. Now it seems that some of the characteristic of this data work is that it's hidden from us even in ways that are more conspicuous than before. I mean programmatic is already over complexity which is so high that it's very hard to track all this data flow the different stage of data work. But it's still can sort of been done right? There are you can investigate I did it protocols or data object specification. You can still have an idea what's going on in terms of data work behind the ecosystem. What about this AI technology, what kind of access we will have into this data work. How can we Account for what's going on, this decision making embedded in data work. These are questions that are kind of new or at least these models they posit in a kind of different scale, I would say. And they are urgent questions because we cannot account for a process if we cannot see anything about that process. And this of course has some consequences and implications.

Speaking of implications and urgent questions as well. You know, one thing that the book touches upon quite a lot is the concept of digital surveillance, which is something that that's been really spoken about in the past 10 years, especially with social media. The Shoshana Zubov book the Age of Surveillance Capitalism is a really great book on this topic. Your book proposes different perspective on digital surveillance. Right. So oftentimes the narrative on surveillance is that there is this big Brother in the form of large tech companies observing and monitoring the digital exhaust we leave in the world. Maybe could you expand on how you view digital surveillance and the surveillance issue and how you tackle it in the book?

There are many problems linked to data, data work and these big platform companies and certainly the idea that they track user behavior and the extract value of it is a very complex one. Now the current idea of surveillance to us is very much based on a means end relation. They do that because they want to surveil us kind of thing, which is like the big Brother, let's say, allow Orwell. But once this is a little bit limiting and limiting on the side of users, not on the side of the company. And I will explain why. If you put user in this sort of double bind in these things that they cannot go out from, because either you are surveyed or you have no choice, then there is very little that we can make as society. But if you instead start looking at surveillance as a sort of byproduct or something more complex that involved also the participation of users. I mean, we are not stupid. Like at least I, you know, right? We want, we go there, we use this technology because we get something out of it and there are services and there are, you know, intelligence machine that answer facilitate our job. They let us do things in different ways, they make us company, whatever. So then it becomes a little bit more complex because surveillance is not any longer the means to an end. I'm not being surveyed because I want to be surveyed, because they want to survey me. Because there is a more complex, let's say production change that involves also the tracking of user behavior. And if we look at it at this with good and bad, we don't want to discount the bad. Let me Be clear, we're not saying, oh guys, that's all very beautiful, there's nothing going on here. No, but we are saying we cannot go backward, we can only move forward. We have these technologies and we want to use them. So how do we instead account for these practices of tracking and regulate them, but still allow user to do what they do and what they wanted to do through these technologies and this company to work because we want them to work, otherwise we won't have these services. So then our idea is to open up this deadlock, let's say between surveillance and surveys, and instead try to understand a little bit more the positive thing and protect user and user rights, but also user access to service at the same time regulate and govern other practices that have been, and we know now not, so let's say respectful of user privacy, fundamental rights, manipulation of user behavior. This should not be allowed that for sure. We are not discounting any of that. But perhaps we say let's give also to user a more active role in this and try to find out how can it be even more active with the participation also of governmental agencies and regulation and even a user bottom up civil society that also sort of participate and collaborate in solving, while solving at least, but making this issue bearable, more malleable for our society.

Maybe on a tactical level, if you can provide some examples, how do you view that working out in practice? We're not going to be able to solve the issue of digital surveillance regulation today, but I'd love to see how does that translate to real world examples of how we can improve the state of digital surveillance regulation.

This, of course a very big question, so I'll try my best sort of to think through. So what the book, for instance may suggest is opening up a debate on these themes, because that's what we want to do in the end. You know, the idea that data are not just input and therefore they are not just mean to survey, they are not tracking devices, but much more. It means that if we translate these ideas and we apply it also to this problem of surveillance, we sort of opening up this problem of surveillance and we start asking question of what kind of data, for instance, instance, or what kind of practice linked to this data lead to this or what kind of data stage we can regulate in order to avoid the pitfall of this and that. So regulation is the first and governance and regulation is the first idea that I should say it's very much needed and perhaps we are already late, we are moving, perhaps in a very disorderly way, but at Least we are moving to address this issue. But here is the thing. If we regulate data as tracking device, we don't go that far. The remedy may be worse than the cause. Let's say if we regulate data in this narrow sense, because they aren't, they aren't only tracking device, they are much more than the social science of data, the data lens perspective. Let's see exactly where it needs to be regulated and how by accounting for all these different links of data with society. Remember the everyday life of people and institution as well. And let's open up a debate on how can we do that without throwing away also what we have conquered, the positive side, without risking giving even more power to this big platform, because that may be also a risk involved in regulating badly data. But a second very important point I would say is data literacy. So the literacy and this is the kind of thing we should start talking a little bit more about. I'm not talking about giving at school coding classes, which may be useful, perhaps not that much now with AI, because we know this sort of will be also taken up and automated. But data literacy, what's data? The discussion we are doing right now, what's their link with society? How have I changed my way of thinking because of data? Data literacy, thinking through data in school, making it a matter of debate, of questioning, of inquiry. And I think this is another very important thing that we can do to avoid this sort of reductionist view that paralyzes society, this sort of deadlock of bad and good. We don't want that. We want to grow and decide what kind of society we want and work towards it. And we need the participation of all to do that. That's why data literacy, participation is another key word. Who is going to participate in this debate of data regulation? Only the company and the regulators. Perhaps we need involvement of other dimension of, you know, society, you know, civil society of. So these things are connected though, because civil society will not participate if this discussion is not facilitated in the first place. And a different idea of data does not penetrate society, you know, which means academia, which means media, journalists, which means show like this, which means many other things.

Yeah, I couldn't agree more on that last element, especially on data literacy, because you want a citizenry that is engaged in the conversation. And here data literacy is an essential lever to be able to get people engaged. Specifically because another aspect of the book is that you talk about the blurring of life's boundaries because, you know, you mentioned that data has become such a central point of our lives that There has been a blurring of boundaries between the different spheres of our lives. You know, for example, the economic and social spheres, the work and private spheres of our lives. Right. So maybe walk us through your thinking here in a bit more depth. And why is it data that is foundational to this blurring of the spheres?

Absolutely, and that's a very important point because it's like perhaps more thorough reading of what someone would call surveillance. Because it's this that we are talking about. Our social interaction has become the resources through which economic value is produced. But this is a huge thing. It's not that because I'm tracked then. So what happened when this is the case? What happens to our society if I engage in conversation and this conversation is the basis for. For the production of economic value, what does that mean? How can I still maintain the basic, the fundamental dimension of society, for instance, worker, employees and employer or contractual relation, or the division between work and private life with private life and public life. All of this is cast as data, allow us to do many things in similar way and to produce value out of it. I mentioned before buying groceries, remember? And reading or studying or learning, or even entertaining and learning. So all these things, when they are done through digital devices, and this is the answer to the second part of your question, how data are done, the basis of this blurring of boundaries. When these activities and practices are done through digital devices and via data, they become there is a sort of homogeneity in them, right? Or we start doing things in similar ways. They are all recorded in data or represented via data. And as data represent so many things in the same media, with the same methodologies, techniques, technologies and process, it means that all these things that were also separated in their representation, they can now be combined to come up with new things. An example perhaps may work because that's also very interesting in the sphere of institution. These two, micro and macro, that we always sort of consider in our book. If we think about everyday life, we immediately perceive it because we all live through it. We work at home, you know, we study on the telephone. So there are all this reshuffling. But if we think about institutions, economic institutions, and there's also the other part of the book, the implication of this blurring of boundaries for our economic institution, how can we understand this? So an example is perhaps the work of TripAdvisor, or if you think about any sectors of the economy that before was a very well defined sector, was rather operating within the boundaries of a sectoral knowledge hospitality industry, which meant they were Doing just some things, booking of hotel or selling of holiday packages or whatever. And now all of a sudden, because of this possibility of extending the knowledge and recording action and interaction they can leverage on a number of things, they were not belonging or do not belong traditionally to the hospitality industry. So they can provide entertainment, they can provide restaurant booking, they can provide, you know, services in other content making or service in other areas like Uber or, you know, so also the institution change because of the blurring of boundaries, not just our everyday life and this dominance of platform and ecosystems somehow respond to that. They are better fit to deal with this sort of heterogeneity of knowledge that can be redrawn, let's say, and used to provide services that do not any longer belong to one specific sector as we understood it.

Yeah, well, that's really fascinating. And you know, as we close out our conversation to loop back, you know, what is the path forward on regulation? Right. Especially in a lot of ways, we're moving to a paradigm where machines are actively participating in the datafication process and where AI can really accelerate a lot of the dynamics that we've been talking about today. What do you think is the path forward on regulation?

Look, there are some stuff that need urgent attention from regulators. And I think that is already something that we should discuss. I mean, we cannot regulate everything at the same time. And sometimes, specifically looking at the European Union, you have the feeling that we want to regulate everything at the same time. But certainly there are some urgent matters to address. One is AI, because of course, and I was mentioning before, this use of data without telling us what kind of data, how are they used, Da, da, da. I think this should be solved and should be solved by governance and regulation. I mean, it's very relevant, as I mentioned before, if we look at, AT data as a complex artifact that allow us to understand the world in certain ways, that allow us to understand ourselves to do things. We cannot accept the issue that we have no accountability whatsoever on how data are made, processed, manipulated and used. Then we need to look into that and to regulate that and to govern these processes. I think. So that's the most urgent issue here. And of course there are then different approaches to the regulation of AI and the different ways in which AI application then are packaged and for instance, the different risks that they pose. I think this is also a discussion that should be opened up. Another perhaps important issue that is not that much debated and should be, I think, debated more is the whole discussion of ecosystem that we made, which belongs also to the AI and That was your question. We are seeing all this ecosystem of data production changing because of the advent of AI. So how do we look into that? How do we make sure that there is enough access for economic actor, that there are practices that are in line with what we need or what we want in terms of market practices or production practices. So this is a whole discussion around the ecosystem AI that perhaps needs to be opened up. So I think these three aspects that I mentioned, the data work and the discussion and governance and regulation around data and data work in terms of AI. The discussion about the risks that certain, you know, AI products or application house for certain perhaps also disadvantaged communities, that's also very important. Some community are much more risks than other for certain AI application. And third, the whole discussion around the ecosystem of production of this AI, which has huge bearing in terms of what decision are made, why those decisions and not other. As we mentioned for you know, the case of data. I think these are three interesting aspects aspect. Somehow we have started already debating them. But what the book advocates I think is to open up a little bit more the discussion around this area considering data not as just narrow input that needs to be sort of either siloed or shared or whatever, you know, or ported or not, but rather think about data governance regulation by thinking of data as complex cognitive artifacts that are at the center of the making of our institutions, of our way of seeing the world or our way to communicate with one another. If we do so, I think really we can, and this is my wish, tackle or address complex issues in complex ways. So the complex issue of regulation in a complex way, not by simplifying and risking a simplified approach to it, because it's a very, very, very important issue for everyone and for our society.

Yeah, I couldn't agree more. And I think this is a great way to cap off our episode. Christina, before we wrap up, Christina, any final call to action to share with with the audience?

Read the book. I think that may be one very sort of instrumental call to action, but there it is, it's already out. And by the way, we have this fantastic image on the COVID She's a music, electronic music composer and artist. So we have very happy also about that. But no, I think the call to action is very much around the issue with the discussed so far. So this idea of data literacy opening up a discussion around data that is not limited to data as an instrument or technology, technological instrument, the input to device. We don't need to school people on how to code. If we don't give also the instrument, the space, the capability to discuss the more fundamental aspect of data that we've been talking through this episode. I think that's a very important call to action, which then also links to a different idea of data and perhaps also of society. We want an open society, we want to progress, we want to have our technologies in place, but also we want not to risk, to delegate to technologies important to decision on what we want to be as society. And I think that's very much around. How do we define data? Definitions are important. How do we think about data? So in the book, we advocate for a social science of data. So that's my wish, that we can open up the debate around social science of data with the participation of many.

Yeah, I couldn't agree more. Hopefully, hopefully soon we will have a codified social science of data discipline. But in the meantime, we're doing the best we can. Thank you so much, Christina, for coming on the show.

Thank you so much for inviting me. It's been a big pleasure. Thank you so much for your questions and for engaging with me in a conversation about data rules.

Likewise. Thank you so much.