Alessandro Bozzon and Geert-Jan Houben on the relationship between AI and design
News – 8 December 2022 – Web redactie Kommunikation
“Our mission is to develop technology that is used by people and integrated into everyday life, making the world a better place. We want to train responsible engineers. This can only be achieved by engaging in a very literal conversation with all the various stakeholders who will use that technology. This is exactly the area where methodologies from the world of design and design can help.” A conversation with Geert-Jan Houben, vice-chancellor AI, Data and Digitization and head of the AI initiative and Alessandro Bozzon, professor of Human-Centered Artificial Intelligence at the Industrial Design Engineering (IDE) faculty at TU Delft. In his work, the latter is committed to improving well-being and promoting inclusion through personal social computing systems.
Alessandro, what is your personal motivation for doing this job?
Alessandro Bozzon: “Geert-Jan and I grew up in a different time. We have experienced the transition from the analogue world to the digital world. We were there at the dawn of the digital society. I’ve been playing around with web technology since the beginning, out of curiosity, what it was. We have also witnessed the emergence of intelligent systems in society and we have seen how these systems have changed people and how they have changed society as a whole. Digital technologies hold immense power and we cannot deny how they can also be disruptive, how they can even cause harm. It is very important that we learn to control them.”
Can you explain what your research is about?
Alessandro Bozzon: “My personal interest has always been in the human side of this movement. I wanted to research the human side but be well-informed by technological insight. So I’m interested in the technical elements of AI, but also how they manifest in relation to people using AI. You need to understand—especially in the case of AI—who these people are, what they do, what they want, how they want it, and in what context they want to use the technology. Only then can you create AI technology that can be used, which really helps us, without all sorts of unintended side effects.”
When you say: you have to speak very well to people and understand people, does it sound almost like psychological or sociological research?
Alessandro Bozzon: “It is an interdisciplinary way of doing research, we actually use knowledge that has been created by other disciplines. And knowledge created by colleagues who work on the applied side of AI, such as within the themes of Health and Mobility.”
How should we see it: designing AI systems?
Alessandro Bozzon: “When you hear the word ‘design’, don’t think of aesthetics in this case: of chairs and tables. AI systems do not work in isolation. They are part of socio-technical systems, organizations, of society; it’s about the interaction between the people, the algorithms, the data used to train those algorithms that comes from the people, the interfaces—all of that. And we must shape that.
When we think of engineering, we think of a straight line, from data to model to user. Or from requirements for software to user. In practice, this is not a straight line, but a circle. The design evolves over time as we get a better understanding of people, a better understanding of technology, it keeps changing.”
Geert Jan Houben: “Exactly, if everything goes well, there will be an interaction between software, data, technology and people. And that interplay must be designed carefully and meticulously. I think that’s the gist. Our goal is to ensure that the interaction between the automated part and the human part works well; carefully and effectively.”
Alessandro Bozzon: “When it comes to the relationship between people and technology, we often talk about trust or about adaptation, how we successfully integrate technology into everyday life. What we need to find out is how people can and will use and interpret the technology. This is where design can come into play; to understand first: what is the problem that the technology is trying to solve? And then to see: How can we make that technology succeed? For me, success means not only working as intended, but also truly integrated into everyday life. Sometimes you see AI technology having an effect that was not foreseen: issues of prejudice, discrimination, harm: they all arise from a misalignment between the desired effect and the actual manifestation of the technology. This is exactly where design comes in: to help imagine the future. And we want to involve the people who will use the technology in the design and engineering process.”
This is exactly where design comes in: understanding the problem the technology is trying to solve, and then making the technology successful
Can you give an example?
Alessandro Bozzon: “The Alexa virtual assistant is a classic example of AI technology that has become normal in our daily lives. Ethical principles come into play here: We want Alexa to do no harm, to be helpful, honest, useful. There is a technical aspect: developing the right algorithms. But there are also design questions: What data are we going to use, whose data are we going to use? What will the interaction between the algorithm and the person look like? What kind of conversations will be held? Which not? It are all things we ‘shape’.”
Geert Jan Houben: “On the one hand, it is important that this type of technology is properly understood by people: it is about the right interpretation. And on the other hand, the form is not always the same either, because it depends on the context. So I think meaning and context are two very important elements.”
Alessandro Bozzon: “Many AI projects and innovations are developed, which then never find a good context in society, there are lots of inventions that no one uses or wants. Because no effort was ever made to understand what was really needed. If you’re developing an AI innovation for a factory, you want people on the shop floor to be part of the design process from the start. I am working on a Horizon project called COALA on this. We create technology through the eyes of the people who will use it; we understand their level of education, their fear of technology and so on. This is the basic principle of what we mean by AI in Delft: AI is not a technology to replace people, but to support people.”
Geert Jan Houben: “The word adaptation is important. AI. Saying “use” gives the impression that once you’ve decided how you want to use it, you just need to create it. But in the case of AI, it is an interactive and continuous process. Our understanding of people is evolving, our relationship with technology is also evolving with lightning speed. It is always evolving.”
In the case of AI, there is an interactive and continuous process. Our understanding of people is evolving, our relationship with technology is also evolving at lightning speed: this is always evolving.
Geert Jan Houben:
If you design an agricultural machine, surely this also applies?
Alessandro Bozzon: “I would like to say that AI is a revolutionary technology that we have never seen before. Revolutionary because AI technology changes because of the way we humans interact with it. There is a deep co-dependency relationship: the way we use technology to change it. Think of filter bubbles or polarization that you see in the world now. Was it possible to predict in advance that social media algorithms would have this effect? Maybe. But above all, it shows that it is impossible to consider an AI innovation ‘finished.’ In that sense, it is fundamentally different from designing an agricultural machine.”
Geert Jan Houben: “You design a solution to a specific problem, then individuals start using it, and they are also influenced by their surroundings and by others; there are all sorts of processes and effects at play, and we see them reflected in the technology that adapts to that use.”
Alessandro Bozzon: “There has always been a human-technology relationship, but the difference with AI systems is that, unlike other systems, they do not always exhibit behavior that can be predicted in advance. Effects may occur that were more difficult to predict. And they are dynamic. They affect us personally, they affect the society we live in. No other technology has changed humanity and the world as quickly and as intensely as the web and data-driven intelligent systems.”
Alessandro Bozzon: “The design perspective is about seeing people as they are, not as you want them to be. And yes, people have different technical abilities, different levels of education, and so on. Design can enable customization without people having to think about it. Good technology is technology that disappears, that just melts away.”
Geert Jan Houben: “With good design, you can make the user focus on the actions and the actions he or she wants them to perform. That the user is not burdened with other things. With the aim of trying to get the user to make good decisions.”
And you can make this here at TU Delft, technology that merges and that people can do good things with?
Alessandro Bozzon: “The great thing about TU Delft is that researchers work together, which I believe is necessary to develop and design the desired AI technology. Delft AI labs is a good example of this: the technological perspective, the ethical perspective and the design perspective come together in AI labs, where the dialogue is conducted. This is actually what society asks of us. And our students learn that too. Not only do they learn to work together, but they learn to understand what it takes to create AI and to create systems that will have a positive impact on the world.”
In the AI laboratories, the technological perspective, the ethical perspective and the design perspective meet, where the dialogue is conducted
Also read the interview with Geert-Jan Houben and Jeroen van den Hoven about developing and maintaining values in AI and in a digital society.