What’s a psychiatrist doing in computer science? - English

At yet another welcome event at Harvard, I get ready to repeat my introduction again. In short: I am a psychiatrist from Belgium, and I will be working for a year as a postdoctoral researcher in computer science. Then I anxiously wait to see if that one question returns: “What’s a psychiatrist doing in computer science?”

OK, it may be unusual, but let me explain why it is a must for me. Computers and the internet have become inseparable from our lives. I still remember clearly when we first got the internet at home: suddenly a whole world opened up, even if it was filled with slowly loading websites competing with each other for bad color schemes and strange fonts. Many years later, AI (artificial intelligence) has entered the stage. Algorithms can learn all sorts of things we instruct them to do, from recognizing license plates to suggesting music.

Over the years, I have become passionate about AI. Still, the first time I ever coded was terrible. I felt anxious and thought that, with my background, I would never be able to do it. But step by step, error after error, coding started to go more smoothly, and it became something I enjoy. From programming psychological tasks to building predictive models for eating behavior and alcohol use in daily life. AI is remarkably good at predicting human behavior and at influencing it. This is exactly why TikTok is so successful at keeping you doomscrolling, even when you don’t know why yourself.

And that is the dark side of AI. Algorithms have never taken a course in ethics. They blindly do what we instruct them to, and if you ask them to absorb as much user attention as possible, that is exactly what they will do. It is therefore our responsibility to handle AI in a responsible way. And that is why it is a must for me to conduct research this year at the department of computer science. I work on algorithms that can help people quit cannabis use or stick to their diet more easily. Imagine that we could follow people in their daily lives through a smartphone app, where an algorithm decides whether help is needed and, if so, what kind. That way, we could support people precisely when they need it most, and in a way tailored to their needs.

In short: just as computers and the internet have become indispensable in our lives, AI will soon be firmly embedded in our world as well. It has enormous potential to help people move forward, but only if we steer it in the right direction. Right now, we can still shape it. But if we do nothing, we should not be surprised if it turns into a monster.

“That’s why we need psychiatrists in computer science.”